How to use the English Dictionary PROPERLY to find Statistical Terms
Reef Fish
to be ridiculed by Richard Ulrich and others who MISUSED the
English Dictionaries.
If one wants to know the definition or meaning of a term in Statistics,
such as the "mode", the best source would be a statistical textbook
that defines it. But if one resorts to the use of an English
Dictionary,
then one would look up the word MODE and then find (among its
many meanings) the definition for Statistics or Mathematics.
I never felt the need before to illustrate that very simple idea that I
had stated numerous times.
This time, I'll play the English Dictionary game and show Richard
Ulrich and others how it should be used properly! In each of the
examples below, I give the source (URL) of the online dictionary as
well as the source of the actual definition for the word "mode":
Merriam-Webster Online Dictionary
==========================
http://www.m-w.com/dictionary/mode
7 a : the most frequent value of a set of data
b : a value of a random variable for which a function of
probabilities defined on it achieves a relative maximum
The American Heritage Dictionary
=========================
http://www.yourdictionary.com/
http://www.yourdictionary.com/ahd/m/m0361900.html
6. Statistics The value or item occurring most frequently
in a series of observations or statistical data.
http://dictionary.cambridge.org/
=======================
http://dictionary.cambridge.org/define.asp?key=99873&dict=CALD
mode (MATHS) Show phonetics
noun [C] SPECIALIZED
the number or value which appears most frequently in a particular set
MSN Encarta Dictionary
==================
http://encarta.msn.com/encnet/features/dictionary/dictionaryhome.aspx
Definition:
6. mathematics statistics most frequent value: the value that
has the highest frequency within a statistical range
http://www.wordcentral.com
====================
Main Entry: 1mode
4 : the most frequent value of a set of values (as data)
Of course, any English dictionary, online or not online, would have
given the SAME, unmistakable meaning/definition of the word MODE
as it is used in Statistics (or Math). It then becomes completely
unequivocal (from "most frequent value" and "value that has the
highest frequency") that a statistical "mode" is NOT an Average of
any kind.
Why did Richard Ulrich and others err, and erred so badly? They
MISUSED the English dictionaries and other sources on the web of
course.
Now let's take a closer look at Richard Ulrich's fallacious arguments
in action (always accompanied by a pinch of his ad-hominem attack
on my inability to read <G>). All of the quotes attributed to Ulrich
are in the PRESENT, ongoing thread,
"A statistical Mode is NOT an Average" Revisited
I welcomed Richard Ulrich's appearing in the thread,
RF> I was expecting you to meet the challenge I had put forth for you
[snip] Richard modestly wrote
RU> And a damn fine job I did of it, too!
RF> But, if Richard Ulrich had learned how to use a dictionary to look
RF> up the word "MODE" whose meaning is at stake, would he have
RF> found any thing other than the "most frequently occurring value"?
I never felt the need to actually exhibit those readily available
definitions
in English dictionaries about the word "MODE", but I felt compelled to
do it this time to make the point emphatic, with clear ILLUSTRATIONS,
as in the definitions at the beginning of this post.
This is Richard Ulrich's typical response, with a little insult thrown
in:
RF> Well, yes. Didn't Bob? Reading deficiency?
Ulrich was of course referring to his MISUSE of the English dictionary
(by not looking up the word "mode" but into all kinds of irrelevant
tangents on the internet!)
RF> Then even HE could have worked out that "A mode is NOT an
RF> average".
This was in reference to the definitions of "mode" shown above that
I hadn't actually exhibited before.
Ulrich> Bob again is arguing from "nothing", that is, the fact that
Ulrich> he reads one reference that omits it; and he throws
Ulrich> away all the references that disagree with him.
Now I challenge Richard Ulrich to find an English dictionary reference
when he looks up the word "mode" that DISAGREES with any of
the references I cited from English dictionaries that ALL pointed
clearly to the FACT that
"A statistical Mode is NOT an Average"
Ulrich> > Here is how dictionaries work -- Some are more complete
Ulrich> > than others.
This is a good stopping point for this segment. I have just given
Ulrich a lesson on how the English dictionary works when anyone
wants to find out the meaning or definition of the word "mode".
FIVE English dictionaries on "mode" cited above -- all gave the
unambiguous, unanamous meaning of a statistical mode, and a
100% clear implication that it is NOT any kind of "average".
-- Reef Fish Bob.
Richard Ulrich
it answered all the questions that he raises while objecting
to it. I think this is the key -
> Ulrich was of course referring to his MISUSE of the English dictionary
> (by not looking up the word "mode" but into all kinds of irrelevant
> tangents on the internet!)
Here is that "tangent" : When you Google for just one word,
Google puts the word "definition" up there beside the count,
to offer a page with definitions. Click on the word "definition."
If you google for <mode average>, Google writes both words
with underlines, and you can click on either, to get those
definitions. There are a dozen or so named sources, for these.
I kind of thought that everyone knew about and used that.
So, here, I am repeating some of the post, with more detail,
since Bob could not work that out.
On 18 Jul 2006 11:08:04 -0700, "Reef Fish"
<Large_Nass...@yahoo.com> wrote:
> This is an explicit illustration of what I said many times before, only
> to be ridiculed by Richard Ulrich and others who MISUSED the
> English Dictionaries.
>
> If one wants to know the definition or meaning of a term in Statistics,
> such as the "mode", the best source would be a statistical textbook
> that defines it. But if one resorts to the use of an English
> Dictionary,
> then one would look up the word MODE and then find (among its
> many meanings) the definition for Statistics or Mathematics.
>
> I never felt the need before to illustrate that very simple idea that I
> had stated numerous times.
>
> This time, I'll play the English Dictionary game and show Richard
> Ulrich and others how it should be used properly! In each of the
> examples below, I give the source (URL) of the online dictionary as
> well as the source of the actual definition for the word "mode":
>
[snip, dictionary citations.]
>
> This is Richard Ulrich's typical response, with a little insult thrown
> in:
>
>
> RF> Well, yes. Didn't Bob? Reading deficiency?
>
> Ulrich was of course referring to his MISUSE of the English dictionary
> (by not looking up the word "mode" but into all kinds of irrelevant
> tangents on the internet!)
Here is where he got lost on what he thought were my
"irrelevant tangents on the internet."
In any case -- I was referring to the
1) Google-definition that said that a mode is an average, which
I quote down below, after Bob's "challenge"; and to
2) another post (by googleUK or something) that gave a reference
which Bob dismissed as a lousy statistical reference; and to
3) another Google-definition item that gave the citation to
Average where its own definition was made, that an average
is a median or mode.
Doing the Google-definition with "average" gives three sources
that explicitly list the mode as *a* definition of average, plus the
Wikipedia reference which fuzzily seems to include it.
I mentioned all those, last time.
[snip]
>
>
> Ulrich> Bob again is arguing from "nothing", that is, the fact that
> Ulrich> he reads one reference that omits it; and he throws
> Ulrich> away all the references that disagree with him.
>
> Now I challenge Richard Ulrich to find an English dictionary reference
> when he looks up the word "mode" that DISAGREES with any of
> the references I cited from English dictionaries that ALL pointed
> clearly to the FACT that
>
> "A statistical Mode is NOT an Average"
>
> Ulrich> > Here is how dictionaries work -- Some are more complete
> Ulrich> > than others.
The lines marked Ulrich seems as true today as they were yesterday.
Not only are some dictionaries more complete, but - in the
light of Bob's selective list - one dictionary *will* have
definitions that may be vastly different, just because those
usages do exist. (Posters in Alt.usage.english occasionally
mention words that are their own antonyms.)
- As to the challenge -
Columbia University Press -
mode, in statistics, an infrequently used type of average. In a
group of numbers the mode is the number occurring most frequently.
In the group 1, 4, 5, 5, 6, 6, 6, 6, 9, 9, the mode is 6 because it
occurs four times and the others only once or twice.
Oops! That says "encyclopedia." Close enough?
Close enough for me. See the others, too.
>
> This is a good stopping point for this segment. I have just given
> Ulrich a lesson on how the English dictionary works when anyone
> wants to find out the meaning or definition of the word "mode".
>
> FIVE English dictionaries on "mode" cited above -- all gave the
> unambiguous, unanamous meaning of a statistical mode, and a
> 100% clear implication that it is NOT any kind of "average".
Further.
Just about every dictionary will equate "average" to "typical".
That seems to me to offer "implications" that Bob is blind to,
or he would not say "100% clear...." It only takes one black
swan to prove that "not all swans are white", where an
infinite list of white swans is only suggestive of the opposite.
Bob cherry-picks his data, and extrapolates on what he selects.
That is perfectly "logical" in a narrow way. However,
when you select definitions which say Average means "typical"
and ones which list modes and medians, and select the ones
which say a mode *is* a mean -- then the opposite conclusion
is perfectly logical.
Simple examples of logic based on selective data gives
contradictions. That should be no surprise.
People with arguing skills try to use the whole set of data.
So far, we are near the start, where Bob picks-and-chooses
his dictionary definitions. But now, it seems like he denies his
arbitrariness, where I thought at the start that he recognized it.
--
Rich Ulrich, wpi...@pitt.edu
http://www.pitt.edu/~wpilib/index.html
Bob Ling
> Bob had trouble understanding my post of yesterday, since
> it answered all the questions that he raises while objecting
> to it. I think this is the key -
>
> > Ulrich was of course referring to his MISUSE of the English dictionary
> > (by not looking up the word "mode" but into all kinds of irrelevant
> > tangents on the internet!)
>
>
> Here is that "tangent" : When you Google for just one word,
> Google puts the word "definition" up there beside the count,
> to offer a page with definitions. Click on the word "definition."
Why should any mathematician accept the fact that the English language
is not constrained by the rules of mathematics? If I ask you how you
are today, I expect an answer of either 0 or 1.
The dynamics of the colloquial use of the English language has
developed a very rich vocabulary, but with subtle nuances throughout:
they've caused illogical dilemma in which there is a lack of
reciprocity, because definitions of Average do point to Mode/Modal, but
not the reverse, which is abhorrent to the purity of mathematical
logic.
> > "A statistical Mode is NOT an Average"
But we cannot say "An average can not be a mode", based on currently
published English dictionaries. Such as by Googling the command:
define: average (for it returns mode/modal citations)
> > FIVE English dictionaries on "mode" cited above -- all gave the
> > unambiguous, unanamous meaning of a statistical mode...
And yet none of them had the statistical mode definition as their first
listing for mode. They were 7th, 6th, 2nd, 6th, and 4th, respectively.
> It only takes one black
> swan to prove that "not all swans are white", where an
> infinite list of white swans is only suggestive of the opposite.
>
> Bob cherry-picks his data, and extrapolates on what he selects.
Which casts doubt on the validity of all of that Author's published
(and co-published) works.
> So far, we are near the start, where Bob picks-and-chooses
> his dictionary definitions. But now, it seems like he denies his
> arbitrariness, where I thought at the start that he recognized it.
Deny everything. It creates better flames.
Reef Fish
Richard Ulrich wrote:
> Bob had trouble understanding my post of yesterday,
Your post yesterday, and all previous times, and since the FIRST time,
April 2005 were talking about using the English DICTIONARY to argue
that a mode IS an average.
I had repeatedly said you MISUSED the English DIctionary, and so I
showed you (and everyone else) the correct way to use the English
dictionaries to see that
A statistical Mode is NOT an "Average"!
>
> > Ulrich was of course referring to his MISUSE of the English dictionary
> > (by not looking up the word "mode" but into all kinds of irrelevant
> > tangents on the internet!)
>
>
> Here is that "tangent" : When you Google for just one word,
That's right -- it's your irrelevant tangent! Google is NOT an
English
dictionary. And if your Google skill is 1/100 of what you claim you
have, you would have found the correct definition of the "mode" (as
I did and showed in other posts) and that would have been the end
of it too.
> So, here, I am repeating some of the post, with more detail,
> since Bob could not work that out.
Don't bother.
You are the ONLY person left in these SCI.STAT groups who
still cling to your ERROR about the meaning of the statistical MODE
(or how to ascertain what it is)!
It's bad enough for my former "tree stump" undergraduates to have
occasionally missed the definition and application of the statistical
term "mode", it is a NATIONAL DISGRACE for someone who calls
himself a statistician, and presumably have taken some legit
courses in statistics, to argue ad infinitum that the statistical IS
an "average", and make all kinds of OTHER errors in statistical
definitions and usages.
> > This is an explicit illustration of what I said many times before, only
> > to be ridiculed by Richard Ulrich and others who MISUSED the
> > English Dictionaries.
> >
> > If one wants to know the definition or meaning of a term in Statistics,
> > such as the "mode", the best source would be a statistical textbook
> > that defines it. But if one resorts to the use of an English Dictionary,
> > then one would look up the word MODE and then find (among its
> > many meanings) the definition for Statistics or Mathematics.
> >
> > I never felt the need before to illustrate that very simple idea that I
> > had stated numerous times.
> >
> > This time, I'll play the English Dictionary game and show Richard
> > Ulrich and others how it should be used properly! In each of the
> > examples below, I give the source (URL) of the online dictionary as
> > well as the source of the actual definition for the word "mode":
> >
>
> [snip, dictionary citations.]
WHY? That was the entire substance of how the English dictionary,
if used properly, would have given EVERYONE the correct definition
and meaning, and not have to bumble around like Richard Ulrich
in his pit of ignorance.
> - As to the challenge -
>
> Columbia University Press -
> mode, in statistics, an infrequently used type of average. In a
> group of numbers the mode is the number occurring most frequently.
> In the group 1, 4, 5, 5, 6, 6, 6, 6, 9, 9, the mode is 6 because it
> occurs four times and the others only once or twice.
>
> Oops! That says "encyclopedia." Close enough?
> Close enough for me. See the others, too.
So, after extensive search, Richard Ulrich cannot find ANY dictionary
that would give a wrong definition of the mode.
Even when he resorted to an "encyclopedia", the CORRECT definition
is there:
> the number occurring most frequently
even with an illustration, for the readership who may be quantitatively
challenged:
> the mode is 6 because it
> occurs four times and the others only once or twice.
Even at that, the encyclopedia assumes that the reader could have
worked out what the ordinary (arithmetic) average IS:
57/10 or 5.7 which is NOT 6, and hence, even the most
unsophiscated person in mathematics and statistics would
have come to the conclusion that
A statistical Mode is NOT an Average
as well as the CORRECT definition of a "mode"!
What does that make Richard Ulrich???
BTW, you FAILED the challenge, miserably, and once more
proved yourself WRONG, by YOURSELF.
>
> >
> > This is a good stopping point for this segment. I have just given
> > Ulrich a lesson on how the English dictionary works when anyone
> > wants to find out the meaning or definition of the word "mode".
> >
> > FIVE English dictionaries on "mode" cited above -- all gave the
> > unambiguous, unanamous meaning of a statistical mode, and a
> > 100% clear implication that it is NOT any kind of "average".
>
> Bob cherry-picks his data, and extrapolates on what he selects.
You are clinging to your ERROR of NOT looking up "mode" in
the English dictionary when the definition of the word wanted
is the statistical "mode".
> --
> Rich Ulrich, wpi...@pitt.edu
> http://www.pitt.edu/~wpilib/index.html
The ignorant, uneducated statistical Quack of sci.stat.math!
-- Reef Fish Bob.
Reef Fish
Bob Ling wrote:
The forger with the posting ID of Large_Nas...@Yahoo.com
had been confirmed by Yahoo long ago (upon my complaint of abuse)
that even the Yahoo address was FORGED (that it didn't exist in Yahoo)
when he also forged the posting name of "Reef FIsh".
This is the first time he forged with the posting name of "Bob Ling",
The profile of " large_nas...@yahoo.com " given by Google:
This person has not created a profile.
large_nas...@yahoo.com
Recent Posts:
Posts in All 6 Groups -- 12 messages
rec.scuba.locations -- 6 messages
sci.stat.math -- 2 messages
sci.stat.edu -- 1 message
rec.travel.cruises -- 1 message
rec.scuba -- 1 message
rec.gambling.blackjack -- 1 message
sci.stat.math 5 days ago
rec.gambling.blackjack July 7
rec.travel.cruises May 9
rec.scuba Mar 13
rec.scuba.locations Mar 2
rec.scuba.locations Feb 20
rec.scuba.locations Feb 18
rec.scuba.locations Nov 21 2005
rec.scuba.locations Nov 15 2005
sci.stat.edu Oct 29 2005
A dedicated FORGER and stalker of Reef Fish Bob whose
LIFE TIME posting record consists of 12 posts, ALL with
forged posting names and posting addresses that made it
appear to be "Reef Fish" Large_Nass...@yahoo.com
> Richard Ulrich wrote:
This is the kind of mentality and poster attracted by Richard Ulrich!
-- Reef Fish Bob.
Brett Magill
<SNIP A LOT OF THE POST>
> Now I challenge Richard Ulrich to find an English dictionary reference
> when he looks up the word "mode" that DISAGREES with any of
> the references I cited from English dictionaries that ALL pointed
> clearly to the FACT that
>
> "A statistical Mode is NOT an Average"
>
> Ulrich> > Here is how dictionaries work -- Some are more complete
> Ulrich> > than others.
>
> This is a good stopping point for this segment. I have just given
> Ulrich a lesson on how the English dictionary works when anyone
> wants to find out the meaning or definition of the word "mode".
>
> FIVE English dictionaries on "mode" cited above -- all gave the
> unambiguous, unanamous meaning of a statistical mode, and a
> 100% clear implication that it is NOT any kind of "average".
>
> -- Reef Fish Bob.
Or, how about some common sense. Ask any elementary educated 11-year old
to caclulate an average and they will sum the elements and divide the
result by the number of elements. Commonly, 'average' refers to the mean.
I googled 'calculate an average' and every site was related to the mean,
including pages on Excel's AVERAGE function.
I recall that the section in my first stats course that covered mean,
median, and mode was referred to as "measures of central tendency" and I
think this is a better reflection. Each is, in some sense, a measure of
the central tendency of the distribution. In the case of the mode, it
might be the "central tendencies".
Reef Fish
Brett Magill wrote:
> On Tue, 18 Jul 2006 11:08:04 -0700, Reef Fish wrote:
>
> <SNIP A LOT OF THE POST>
>
> > Now I challenge Richard Ulrich to find an English dictionary reference
> > when he looks up the word "mode" that DISAGREES with any of
> > the references I cited from English dictionaries that ALL pointed
> > clearly to the FACT that
> >
> > "A statistical Mode is NOT an Average"
> >
> > Ulrich> > Here is how dictionaries work -- Some are more complete
> > Ulrich> > than others.
> >
> > This is a good stopping point for this segment. I have just given
> > Ulrich a lesson on how the English dictionary works when anyone
> > wants to find out the meaning or definition of the word "mode".
> >
> > FIVE English dictionaries on "mode" cited above -- all gave the
> > unambiguous, unanamous meaning of a statistical mode, and a
> > 100% clear implication that it is NOT any kind of "average".
> >
> > -- Reef Fish Bob.
>
> Or, how about some common sense.
That's setting the bar MUCH too high. Remember, I was addressing
Richard Ulrich. Did you see his follow-up? I had appealed to
everyone's sense -- common or not, for over a year in this group,
and Richard Ulrich was still arguing.
So, I lowered myself ot his level of insisting on using the English
Dictionary to learn what a statistical definition is, and played his
game on his home court, to show that he MISSED not only the
GOAL of whatever game he played, but missed the COURT too.
> Ask any elementary educated 11-year old
> to caclulate an average and they will sum the elements and divide the
> result by the number of elements. Commonly, 'average' refers to the mean.
Everyone knows that. But that's secondary to the statement,
"A statistical Mode is NOT an average." In fact many of those who
argued against the statement (in other groups) flaunted their
half-baked learning by citing that the average could be the geometric
mean of the harmonic mean, etc.
But it dosn't depend on the commonsense "arithmetic mean" for
anyone of having SOME SENSE to know that a Mode is NOT
any kind of average.
> I googled 'calculate an average' and every site was related to the mean,
> including pages on Excel's AVERAGE function.
You're playing Ulrich's other game of Gooling for the word "average"
and shot him down even on THAT unnecessary and inappropriate
way of looking for the correctness of the statememnt in question.
>
> I recall that the section in my first stats course that covered mean,
> median, and mode was referred to as "measures of central tendency" and I
> think this is a better reflection.
That is indeed the STANDARD terminlogy in elementary
statistics books which I've taught from DOZENS. It carries
the exactly correct classification of those terms,
> Each is, in some sense, a measure of
> the central tendency of the distribution. In the case of the mode, it
> might be the "central tendencies".
And also the EXTREME non-central value, such as the
minimum or maximum -- which would portray the opposite
extremes of the other two measures of central tendency.
The AVERAGE can also be far from the central value of
the median -- which is why the reporting of the "central
tendencies" of commonly known SKEW distributions,
such as INCOME, total wealth, or other distributions
that are bounded on the left by ZERO, and highly skewed
to the right.
For those distributions, the MEDIAN is used.
Otherwise, for the distribution of assets of residents of
the city of Seattle, Washington, Bill Gates alone would
single-handedly put the population MEAN (average)
of family assets to a point that is exceeded only by
himself. :-)
Thanks for giving your "commonsense" perspective.
I like your 11-year old example. Most of those 11
year olds would not know what a statistical MODE is,
but once told that it's the "most frequent" value (or
values), would immediately know that it's not the
AVERAGE.
"A little knowledge is worse than none" clearly applied
to the "crowd folly" in several newsgroups in which I
had the experience of meeting many Richard Ulrich
clones.
The best thing I can say about sci.stat.math is that
Richard Ulrich appears to be the ONLY remaining
person in our group who remained ignorant,
and has no commonsense whatsoever
relative to the statement
"A statistical Mode is NOT an Average"
and had been shown the correct definition and
explanation by several others in this group.
The BAD news is that we have SOMEONE as ignorant
(and obnoxious) about elementary statistical concepts in
sci.stat.math.
The GOOD news is that we seem to have only TWO
(Luis A. Afonso and Richard Ulrich), and Richard Ulrich
has far surpassed Afonso in his DIVERSITY of
ignorance in statistical terms.
-- Reef FIsh Bob.
Brett Magill
<CUT PART OF POST>
>> Ask any elementary educated 11-year old
>> to caclulate an average and they will sum the elements and divide the
>> result by the number of elements. Commonly, 'average' refers to the mean.
>
> Everyone knows that. But that's secondary to the statement,
> "A statistical Mode is NOT an average."
Everyone knows that? LOL, I believed that the fact that everyone did
not know that was the raison d'etre of this thread.
As for the relevace of my post to the statement, "A statistical Mode
is NOT an average," let me connect the dots.
My point was that average is equivalent to the artihmetic mean. When
asked to calculate an average, there is no ambiguity. It is perfectly
clear in that there is a one to one correspondence between the two. So,
to the degree that average = arithmetic mean, it would be absurd to
suggest that mode = average, which would imply that arithmetic mean = mode.
Intrigued by the argument and surprised that it even existed among anyone
who practices statistics in any capacity, I went looking for places where
it was used incorrectly. I did see a few web pages that used "average" to
refer to measures of central tendency more generally. Of course, even
those make the distinction between mean and mode and usually even
explicitly say mean=average. This clumsy usage would be better replaced
by "measures of central tendency".
Besides, I wanted to throw in my .02 and play a small part in the theater
of the absurd that this thread has become...
Bruce Weaver
--- snip ----
> I did see a few web pages that used "average" to
> refer to measures of central tendency more generally.
I don't have my copy of Darrell Huff's "How to Lie with Statistics"
here, but I do remember that he described the median as a type of
average. At amazon.com, I was able to dig out all references to the
mode (see below), and discovered that he also describes it as a kind of
"average" too.
1. on Page 28:
"... very much about it unless you can find out which of the common
kinds of average it is-mean, median, or mode. The $15,000 figure I used
when I wanted a big one is a mean, the arithmetic average of the incomes
..."
2. on Page 29:
"... idea of the stature of these people. You don't have to ask whether
that average is a mean, median, or mode; it would come out about the
same. (Of course, if you are in the business of manufacturing overalls
for these ..."
3. on Page 30:
"... distribution. If you draw a curve to represent it you get something
shaped like a bell, and mean, median, and mode fall at the same point.
Consequently one kind of average is as good as another for describing
the heights of ..."
4. on Page 32:
"... get how much. The boss might like to express the situation as
"average wage $5,700"- using that deceptive mean. The mode, however, is
more revealing: most common rate of pay in this business is $2,000 a
year. As usual, the median ..."
--
Bruce Weaver
bwe...@lakeheadu.ca
www.angelfire.com/wv/bwhomedir
Reef Fish
Brett Magill wrote:
> On Wed, 19 Jul 2006 10:35:41 -0700, Reef Fish wrote:
>
> <CUT PART OF POST>
> >> Ask any elementary educated 11-year old
> >> to caclulate an average and they will sum the elements and divide the
> >> result by the number of elements. Commonly, 'average' refers to the mean.
> >
> > Everyone knows that. But that's secondary to the statement,
> > "A statistical Mode is NOT an average."
>
>
> Everyone knows that?
This is the kind of misunderstanding arising from the inherent
ambiguity
in the use of the English language, and can easily be clarified.
"Everyone knows that" referred to YOUR three lines, about the 11-year
olds knowing how to calculate the average.
It's my second sentence that needs to be clarfied even though I thought
I had already explained what I meant. Perhaps I didn't. Let's see ...
Yup. IMMEDIATELY after your cited two lines, I wrote,
BEGIN quote,
In fact many of those who
argued against the statement (in other groups) flaunted their
half-baked learning by citing that the average could be the geometric
mean of the harmonic mean, etc.
But it dosn't depend on the commonsense "arithmetic mean" for
anyone of having SOME SENSE to know that a Mode is NOT
any kind of average.
END quote.
What I meant to convey was that knowing what the 11-year old knows
is far from sufficient, because "those who argued against the
statement"
knew what the arithmetic average IS -- and they cited other kinds of
averages, and missed the LOGIC that the statement does NOT
depend on the commonsense meaning of "average" as understood by
any 11-year old. They dragged in OTHER meanings of average to
trip themselves.
That was what I meant by
> > But that's secondary to the statement,
> > "A statistical Mode is NOT an average."
meaning the PRIMARY mistake of the "older crowd" is not that they
don't know what the 11-year olds know about the arithmetic average,
but they FAILED to follow the logic, as ALL 11-year olds would follow,
once told that the 'mode' is the "most frequent" value, to conclude
> > "A statistical Mode is NOT an average."
> LOL, I believed that the fact that everyone did
> not know that was the raison d'etre of this thread.
No, I think everyone DOES know that the mean IS an average.
Hey, even Richard Ulrich knew both of those!
Their crime was caused by:
PRIMARY: The failure in LOGIC after the meaning of the "mode"
(highest frequency) and the mean (the arithmetic
average"
had been explained AND agreed by them.
Richard Ulrich FAILED in this primary reason when
no 11-year
old would have failed.
SECONARY: The commonsense meaning of "Average" is the "arithmetic
average" or the ordinary "mean".
It doesn't really matter WHAT average we are
talking about.
Hence, secondary.
Having explained the above, I do agree that the intended meaning was
not "absolutely obvious". :-) But I think it would be NOW.
That reminded me only of the story aboutt a professor of calculus
having been asked a question that wasn't immediately obvious to
him how to answer. So, he strode outside of the classroom, and
came back later, to give his answer, "Yes, it is obvious."
In the case of the two lines whose meaning escaped you, at least
I used at least 20 lines to explain the unmistakable (intended)
meaning, leaving no tern unstoned.
> As for the relevace of my post to the statement, "A statistical Mode
> is NOT an average," let me connect the dots.
I actually had already connected all the dots, as you should be able
to see from my extended explanation ABOVE.
>
> My point was that average is equivalent to the artihmetic mean. When
> asked to calculate an average, there is no ambiguity. It is perfectly
> clear in that there is a one to one correspondence between the two. So,
> to the degree that average = arithmetic mean, it would be absurd to
> suggest that mode = average, which would imply that arithmetic mean = mode.
You last sentence confirmed what I meant PRIMARY, and even
generalized that to allow other SECONDARY meanings of the word "mean"!
To be explicit, <G>, my GENERALIZED statement contained yours as
a special case:
> to the degree that average = <SOME> mean, it would be absurd to suggest
> that mode = average, which would imply that <SOME> mean = mode.
Kevin: SOME = 'arithmetic'
others: SOME = 'geometric'
still others: SOME = 'harmonic', 'weighted', etc.
Your statement about the "absurdity to suggest" holds.
>
> Intrigued by the argument and surprised that it even existed among anyone
> who practices statistics in any capacity, I went looking for places where
> it was used incorrectly. I did see a few web pages that used "average" to
> refer to measures of central tendency more generally. Of course, even
> those make the distinction between mean and mode and usually even
> explicitly say mean=average. This clumsy usage would be better replaced
> by "measures of central tendency".
That is TERTIARY (3rd), as opposed to SECONARY. :-)
>
> Besides, I wanted to throw in my .02 and play a small part in the theater
> of the absurd that this thread has become...
Indeed! The Theatre of the Absurd is very much alive and well!
Your 19.1 KRW (S.Korean Won) or 0.16 CNY (China Yuan) OR
especially the 3.18 PTE (Portugal Escudos) are greatly appreciated.
http://www.xe.com/ucc/convert.cgi
-- Reef Fish Bob.
Reef Fish
Bruce Weaver wrote:
> Brett Magill wrote:
>
> --- snip ----
>
> > I did see a few web pages that used "average" to
> > refer to measures of central tendency more generally.
>
> I don't have my copy of Darrell Huff's "How to Lie with Statistics"
> here, but I do remember that he described the median as a type of > average. At amazon.com, I was able to dig out all references to the
> mode (see below), and discovered that he also describes it as a kind of
> "average" too.
You mean "reference to the AVERAGE in which the mode is mentioned"
didn't you? But that's PRECISELY what one should NOT do -- one
should ask about the MODE.
If you are using the examples below to convey the FALLACY of usage
by the "Crowd Folly" by looking up the wrong reference WORD --
which should be the "mode", but not the "average". Fine!
Otherwise, you are committing the SAME fallacious arguments (to
support the absurd idea that the mode is an average) that had been
advanced at least a thousand times (and that's counyinh only those
I had seen pr heard first hand). :-)
The plot in the Theatre of the Absurd may have thickened!
-- Reef Fish Bob.
Richard Ulrich
<magillb@*nomail*.sbcglobal.net> wrote:
[snip]
>
> Or, how about some common sense. Ask any elementary educated 11-year old
> to caclulate an average and they will sum the elements and divide the
Now, "calculate an average" is introducing another aspect.
- Bob does admit that the harmonic mean, etc., qualify as
averages, and they fail your common-sense test.
===== start, Bob citing Bob (so it is no accident) .
RF> The "arithmetic" mean, the "goemetric" mean, the
RF> "harmonic" mean, quadratic mean, weighted mean, as
RF> well as trimmed (or truncated) mean, and Winsorized
RF> mean, among others.
RF> All of those means are AVERAGES.
===== end
I'm still wondering who will comment on the "family of means"
that I have mentioned several times -- the one defined by
minimizing loss. The usual mean minimizes squared deviations,
the median minimizes absolute deviation, and the mode minimizes
the count of deviations (in groups or categories).
I took this as a poignant connection between the average and
mode, which should be hard to ignore on the mathematical side.
> result by the number of elements. Commonly, 'average' refers to the mean.
Commonly, it does. The "arithmetic mean" is the modal definition.
The Proposition does not say, "commonly".
> I googled 'calculate an average' and every site was related to the mean,
> including pages on Excel's AVERAGE function.
>
> I recall that the section in my first stats course that covered mean,
> median, and mode was referred to as "measures of central tendency" and I
> think this is a better reflection. Each is, in some sense, a measure of
> the central tendency of the distribution. In the case of the mode, it
> might be the "central tendencies".
"Central tendency" is surely one of the connecting ideas
Richard Ulrich
<Large_Nass...@yahoo.com> wrote:
>
> Richard Ulrich wrote:
> > Bob had trouble understanding my post of yesterday,
RF >
> Your post yesterday, and all previous times, and since the FIRST time,
> April 2005 were talking about using the English DICTIONARY to argue
> that a mode IS an average.
Now THAT is surely the truth, and I've been telling him
that since near the start. Most people don't have the problem,
but *he* can't read me.
[snip]
RU > > - As to the challenge -
> >
> > Columbia University Press -
> > mode, in statistics, an infrequently used type of average. In a
> > group of numbers the mode is the number occurring most frequently.
> > In the group 1, 4, 5, 5, 6, 6, 6, 6, 9, 9, the mode is 6 because it
> > occurs four times and the others only once or twice.
> >
> > Oops! That says "encyclopedia." Close enough?
> > Close enough for me. See the others, too.
RF >
> So, after extensive search, Richard Ulrich cannot find ANY dictionary
> that would give a wrong definition of the mode.
One click, the first place I looked, is hardly an extensive search.
Comments.
1) As references go, why is an encyclopedia inadequate?
2) I figured that I deserved some extra credit for having
answered the question for Bob, a day earlier than the question.
Even though he could not read it. Or, did he see the encyclopedia
citation and want a separate one?
3) Two days ago, another user cited and quoted a dictionary,
which was a statistical dictionary.
Now, what was the point of the Challenge?
Was there an implication that Bob would admit ignominious
defeat?
Naah...
Reef Fish
Richard Ulrich wrote:
> On Wed, 19 Jul 2006 16:40:48 GMT, Brett Magill
> <magillb@*nomail*.sbcglobal.net> wrote:
> [snip]
> >
> > Or, how about some common sense. Ask any elementary educated 11-year old
> > to caclulate an average and they will sum the elements and divide the
>
> Now, "calculate an average" is introducing another aspect.
> - Bob does admit that the harmonic mean, etc., qualify as
> averages, and they fail your common-sense test.
>
> ===== start, Bob citing Bob (so it is no accident) .
> RF> The "arithmetic" mean, the "goemetric" mean, the
> RF> "harmonic" mean, quadratic mean, weighted mean, as
> RF> well as trimmed (or truncated) mean, and Winsorized
> RF> mean, among others.
>
> RF> All of those means are AVERAGES.
> ===== end
> I took this as a poignant connection between the average and
> mode, which should be hard to ignore on the mathematical side.
You are STILL missing the point that NONE of those averages
can be construed to mean "the highest frequency", the mode?
> --
> Rich Ulrich, wpi...@pitt.edu
> http://www.pitt.edu/~wpilib/index.html
A Blue Ribbon Award to Richard Ulrich for his performance at
the Theatre of the Absurd.
-- Reef Fish Bob.
Bruce Weaver
> Bruce Weaver wrote:
>> Brett Magill wrote:
>>
>> --- snip ----
>>
>>> I did see a few web pages that used "average" to
>>> refer to measures of central tendency more generally.
>> I don't have my copy of Darrell Huff's "How to Lie with Statistics"
>> here, but I do remember that he described the median as a type of > average. At amazon.com, I was able to dig out all references to the
>> mode (see below), and discovered that he also describes it as a kind of
>> "average" too.
>
> You mean "reference to the AVERAGE in which the mode is mentioned"
> didn't you?
No.
> But that's PRECISELY what one should NOT do -- one
> should ask about the MODE.
Let me try again. I remembered that Huff called the median a kind of
"average". I could not remember if he also called the mode an
"average". To find out, I searched for every occurrence of the word
"mode". I found the four occurrences shown in my earlier post. In
every one of them, Huff either explicitly calls the mode a kind of
average, or he implies it. (I think that qualifies as asking about the
mode, doesn't it?)
>
> If you are using the examples below to convey the FALLACY of usage
> by the "Crowd Folly" by looking up the wrong reference WORD --
> which should be the "mode", but not the "average". Fine!
>
> Otherwise, you are committing the SAME fallacious arguments (to
> support the absurd idea that the mode is an average) that had been
> advanced at least a thousand times (and that's counyinh only those
> I had seen pr heard first hand). :-)
It took me a minute to work out that "counyinh" was "counting". Must be
time to go home. :-|
I was not defending Huff's usage. I was just pointing it out as another
example of what Brett Magill described as a more general use of
"average" that is more or less equivalent to "measure of central
tendency". Huff's book is fairly widely known, I think, and could be
one of the sources that lead some people to call the mode an average.
>
> The plot in the Theatre of the Absurd may have thickened!
>
> -- Reef Fish Bob.
>>
--
Bruce Weaver
bwe...@lakeheadu.ca
www.angelfire.com/wv/bwhomedir
Reef Fish
Bruce Weaver wrote:
> Reef Fish wrote:
> > Bruce Weaver wrote:
> >> Brett Magill wrote:
> >>
> >> --- snip ----
> >>
> >>> I did see a few web pages that used "average" to
> >>> refer to measures of central tendency more generally.
> >> I don't have my copy of Darrell Huff's "How to Lie with Statistics"
> >> here, but I do remember that he described the median as a type of > average. At amazon.com, I was able to dig out all references to the
> >> mode (see below), and discovered that he also describes it as a kind of
> >> "average" too.
> >
> > You mean "reference to the AVERAGE in which the mode is mentioned"
> > didn't you?
>
> No.
>
> > But that's PRECISELY what one should NOT do -- one
> > should ask about the MODE.
>
> Let me try again. I remembered that Huff called the median a kind of
> "average". I could not remember if he also called the mode an
> "average". To find out, I searched for every occurrence of the word
> "mode". I found the four occurrences shown in my earlier post. In
> every one of them, Huff either explicitly calls the mode a kind of
> average, or he implies it. (I think that qualifies as asking about the
> mode, doesn't it?)
Thanks for the re-explanation and clarification. I DID misunderstand
your explanation the first time.
It was an unusual kind of "look up", not the kind one would do in
trying to find out the meaning of a word. You only looked up when
a certain word appeared and in those instances cited, Huff DID
screw up, like the rest of the "crowd".
Fair enough. I can see now how he can lie with statistics without
trying! :-)
> > If you are using the examples below to convey the FALLACY of usage
> > by the "Crowd Folly" by looking up the wrong reference WORD --
> > which should be the "mode", but not the "average". Fine!
> >
> > Otherwise, you are committing the SAME fallacious arguments (to
> > support the absurd idea that the mode is an average) that had been
> > advanced at least a thousand times (and that's counyinh only those
> > I had seen pr heard first hand). :-)
>
> It took me a minute to work out that "counyinh" was "counting". Must be
> time to go home. :-|
It's my finger-drift that happens more often than I would like to
count.
The three instances of the wrong key in one line were all hitting the
key ADJACENT to the correct one. "counyinh" - y is next to t,
and h is next to g, hence "counting". "pr" -- p is next to o, hence
"or".
Those are more advanced cases of what I call "inconsequential
misspellings" that most people can work out what was intended,
>
> I was not defending Huff's usage. I was just pointing it out as another
> example of what Brett Magill described as a more general use of
> "average" that is more or less equivalent to "measure of central
> tendency". Huff's book is fairly widely known, I think, and could be
> one of the sources that lead some people to call the mode an average.
He certainly has plenty of bad company with dictionary writers on
the entry "average".
Thanks for clarifying what you meant.
-- Reef Fish Bob.
David Winsemius
>
>> --
>> Rich Ulrich, wpi...@pitt.edu
>> http://www.pitt.edu/~wpilib/index.html
>
> A Blue Ribbon Award to Richard Ulrich for his performance at
> the Theatre of the Absurd.
>
> -- Reef Fish Bob.
>
Curtain rises.
The players all take a bow and award themselves Tonys.
The audience shakes off its ennui (1), stumble out of the theater, and
toddle off to bed.
--
From a dictionary of sorts:
(1) Ennui is a feeling of weariness and dissatisfaction.
The phrase mihi in odi est, meaning “I hate or dislike,” gave rise to the
Vulgar Latin verb inodire, “to make odious,” the source of the Old French
verb ennuyer or anoier, “to annoy, bore.” This was borrowed into English
by around 1275 as anoien, our annoy. From the Old French verb a noun
meaning “worry, boredom” was derived, which became ennui in modern
French. This noun, with the sense “boredom,” was borrowed into English in
the 18th century, perhaps filling a need in polite, upper class society.
Ennui is different from boredom in that ennui describes weariness or
annoyance over a long period of time; boredom is relieved easily, while
ennui is constant.
Richard Ulrich
<Large_Nass...@yahoo.com> wrote:
>
> Richard Ulrich wrote:
[snip]
> > RF> All of those means are AVERAGES.
> > ===== end
>
> > I took this as a poignant connection between the average and
> > mode, which should be hard to ignore on the mathematical side.
>
> You are STILL missing the point that NONE of those averages
> can be construed to mean "the highest frequency", the mode?
Bob, are my words invisible to you?
Is that an accident, or an unfair edit?
This is what Bob snipped, without marking it, after "==== end" .
I'm still wondering who will comment on the "family of means"
that I have mentioned several times -- the one defined by
minimizing loss. The usual mean minimizes squared deviations,
the median minimizes absolute deviation, and the mode minimizes
the count of deviations (in groups or categories).
I mentioned at the start, Bob has difficulty with "dialog."
Reef Fish
Richard Ulrich wrote:
> On 19 Jul 2006 15:57:36 -0700, "Reef Fish"
> <Large_Nass...@yahoo.com> wrote:
>
> >
> > Richard Ulrich wrote:
> [snip]
>
> > > RF> All of those means are AVERAGES.
> >
> > You are STILL missing the point that NONE of those averages
> > can be construed to mean "the highest frequency", the mode?
>
> Bob, are my words invisible to you?
> Is that an accident, or an unfair edit?
>
> This is what Bob snipped, without marking it, after "==== end" .
>
> I'm still wondering who will comment on the "family of means"
> that I have mentioned several times -- the one defined by
> minimizing loss. The usual mean minimizes squared deviations,
> the median minimizes absolute deviation, and the mode minimizes
> the count of deviations (in groups or categories).
But what relevant does THAT have on the definition of the MODE
that it is the highest frequency?
That was why I snipped it.
You failed in LOGIC too. It doesn't matter what the mode minimized
or what anything else minimizes, its definition of HIGHEST FREQUENCY
logically implies that it is not, and cannot, coincode with the
DEFINITION
of any MEAN.
>
> I mentioned at the start, Bob has difficulty with "dialog."
You mentioned it again now. Read my preceding 4 lines about your
FAILURE in LOGIC.
-- Reef Fish Bob.
Richard Ulrich
<Large_Nass...@yahoo.com> wrote:
>
> Richard Ulrich wrote:
> > On 19 Jul 2006 15:57:36 -0700, "Reef Fish"
> > <Large_Nass...@yahoo.com> wrote:
[snip]
> >
> > This is what Bob snipped, without marking it, after "==== end" .
> >
> > I'm still wondering who will comment on the "family of means"
> > that I have mentioned several times -- the one defined by
> > minimizing loss. The usual mean minimizes squared deviations,
> > the median minimizes absolute deviation, and the mode minimizes
> > the count of deviations (in groups or categories).
>
> But what relevant does THAT have on the definition of the MODE
> that it is the highest frequency?
>
> That was why I snipped it.
>
> You failed in LOGIC too. It doesn't matter what the mode minimized
> or what anything else minimizes, its definition of HIGHEST FREQUENCY
> logically implies that it is not, and cannot, coincode with the
> DEFINITION
> of any MEAN.
[ ... ]
Okay. My topic seems to concern how the concept
of "average" has widened, with an example that is "legitimate"
or professional, compared to some other historical changes.
You concede that the mode belongs to a "family of
means" that is defined by minimization? But you think
another "family of means" is not relevant?
It seems to me to be parallel to the other family of
means/averages that you concede to be legitimate as averages.
It gets the name, and it is a useful generalization.
Do you insist on not-calling that a family of means
(because it doesn't have to divide by N)?
Or you don't call its members, "means"?
Reef Fish
> On 20 Jul 2006 17:24:07 -0700, "Reef Fish"
> <Large_Nass...@yahoo.com> wrote:
>
> >
> > Richard Ulrich wrote:
> > > On 19 Jul 2006 15:57:36 -0700, "Reef Fish"
> > > <Large_Nass...@yahoo.com> wrote:
> [snip]
> > >
> > > This is what Bob snipped, without marking it, after "==== end" .
> > >
> > > I'm still wondering who will comment on the "family of means"
> > > that I have mentioned several times -- the one defined by
> > > minimizing loss. The usual mean minimizes squared deviations,
> > > the median minimizes absolute deviation, and the mode minimizes
> > > the count of deviations (in groups or categories).
> >
> > But what relevant does THAT have on the definition of the MODE
> > that it is the highest frequency?
> >
> > That was why I snipped it.
> >
> > You failed in LOGIC too. It doesn't matter what the mode minimized
> > or what anything else minimizes, its definition of HIGHEST FREQUENCY
> > logically implies that it is not, and cannot, coincode with the
> > DEFINITION
> > of any MEAN.
> [ ... ]
>
>
> means" that is defined by minimization?
I conceded no such! The mode does NOT belong to any "family
of means". What the mean minimizes is IRRELEVANT to the
definition and meaning of the statistical MODE.
> But you think
> another "family of means" is not relevant?
The "family of means" is IRRELEVANT, to the definition and
proper usage of the MODE. Period.
> Do you insist on not-calling that a family of means
> (because it doesn't have to divide by N)?
> Or you don't call its members, "means"?
I could care less what you call it. We were supposed to be
discussing what a MODE is. And the definition and meaning
and usage of the term MODE has nothing to do with what
any mean, or any member in the family of means.
> --
> Rich Ulrich,
Rich, at this point, it is my opinion, based on your latest posts
about the MODE, that you have entered the realm of being
a statistical TROLL, or you intend to prove that you're not only
obtuse in your understanding of statistics and statistical terms,
but you are INFINITELY obtuse.
If it's the FORMER, I'll play along as being the victim of your
TROLL, for the reason that each time you do so, you lose just
a bit more of whatever little credibility you might have had left
in these sci.stat groups about your statistical knowledge. But
I want to make CERTAIN that you are not misleading some
innocent readers who are new to statistics into following your
ERRORS, your statistical Quackery, and malpractice in
statistics.
If it's the LATTER, may I be the first to congratulate you that
you have given adequate PROOF that you are "infinitely
obtuse". The proof was rather lengthy -- even more so than
the 200-pages it took Wiles to prove Fermat's Last Theorem,
but let me assure you that your proof is complete. Q.E.D.
Was your post a TROLL or a PROOF?
I'll let the readers decide.
-- Reef Fish Bob.
Richard Ulrich
<Large_Nass...@yahoo.com> wrote:
> Richard Ulrich wrote:
> > On 20 Jul 2006 17:24:07 -0700, "Reef Fish"
> > <Large_Nass...@yahoo.com> wrote:
> >
> > >
> > > Richard Ulrich wrote:
> > > > On 19 Jul 2006 15:57:36 -0700, "Reef Fish"
> > > > <Large_Nass...@yahoo.com> wrote:
[snip, some. Considering my lines - ]
> > > > I'm still wondering who will comment on the "family of means"
> > > > that I have mentioned several times -- the one defined by
> > > > minimizing loss. The usual mean minimizes squared deviations,
> > > > the median minimizes absolute deviation, and the mode minimizes
> > > > the count of deviations (in groups or categories).
RF> > >
> > > But what relevant does THAT have on the definition of the MODE
> > > that it is the highest frequency?
> > >
> > > That was why I snipped it.
> > >
> > > You failed in LOGIC too. It doesn't matter what the mode minimized
> > > or what anything else minimizes, its definition of HIGHEST FREQUENCY
> > > logically implies that it is not, and cannot, coincode with the
> > > DEFINITION
> > > of any MEAN.
> > [ ... ]
> >
RU > >
> > You concede that the mode belongs to a "family of
> > means" that is defined by minimization?
RF >
> I conceded no such! The mode does NOT belong to any "family
> of means". What the mean minimizes is IRRELEVANT to the
> definition and meaning of the statistical MODE.
>
RU >
> > But you think
> > another "family of means" is not relevant?
RF >
> The "family of means" is IRRELEVANT, to the definition and
> proper usage of the MODE. Period.
Bob seems to have an odd attitude about classes and members,
and what is apt to be defined, that is inappropriate. Or, he
will not say why mention of the "family" is "irrelevant", when
I think it is, and (I'm pretty sure) other people think so, too.
When only *some* definitions of average and mode include each
other, Bob has wiggle room for arguing that some folks keep them
separate, or ought to --
I further mentioned that if the usage were *controversial*, like
using imply for infer, *some* reference ought to mention it ... but
you have to be familiar with references to be swayed by that.
Consider one other idea/concept -- I looked up Human with
the Google click (which brings up multiple reference books
by using Answer.com). "Human" never mentions Mammal.
The citations for Mammal more often do mention humans.
- Here, the general term included the specific, and not the
opposite. That happens to be the same pattern for average/mode.
Does it matter, which-includes-which?
I expect that that will vary for different concepts.
For the concepts of modes and averages, I am not at all
surprised that the discussions of "averages" are the ones
that are more apt to mention the other. Why does Bob take
the presence as meaningless?
For whatever his reason, Bob has been willing to accept the
family of Generalized means as being real averages, in his
list. The geometric mean, and so on, do not fit everyone's
"common sense" notion of an average, so Bob is departing
from the purely-intuitive, common-man basis ... to that degree.
It might be useful if we had a word that *always* meant
sum-X-divided-by-N, and nothing else, in all contexts.
"Average" isn't that word. But then another word would be
needed for the related *family* where a function gets in there.
(Language may be described with a bit of logic, after the fact,
but even that can't always be done. Some words are - even -
their own antonyms.)
RU >
> > Do you insist on not-calling that a family of means
> > (because it doesn't have to divide by N)?
> > Or you don't call its members, "means"?
RF >
> I could care less what you call it. We were supposed to be
> discussing what a MODE is. And the definition and meaning
> and usage of the term MODE has nothing to do with what
> any mean, or any member in the family of means.
The reason for bringing in "families" is that it points to
potential absurdity. There are reasons that words get used
as they do, and that meanings shift and drift. One
reason is "convenience". Another, which is related, is to
avoid absurdity.
Granting that there are families of means that include the median
and the mode, it seems to me, makes it tough to claim that
the median and mode are not "means." Or averages.
Each of these seems too absurd --
"This is a family of central tendencies, but we
must not call it a family of means."
"This is a family of means, but we never call the
members of it means or averages."
"This is a family of means, which we can call means, but
we must never call these means, 'averages'."
I'm trying to get Bob to endorse one, or to explain why not.
"Avoiding absurdity" is also involved in discussing the
adjective-use of "average" which Bob did not want to allow.
- The best representation of "average income" for purposes
of argument about wages is apt to be the median or mode.
[...]
RF >
> Rich, at this point, it is my opinion, based on your latest posts
> about the MODE, that you have entered the realm of being
> a statistical TROLL, or you intend to prove that you're not only
> obtuse in your understanding of statistics and statistical terms,
> but you are INFINITELY obtuse.
Bob, I keep trying to reduce my arguments to tinier steps,
and stating the assumptions that I assumed were shared,
in hopes that you can grasp them. When you give me
no-comment-whatsoever on them, I don't have much to
work with for articulating further.
If you do, say, *answer* a blunt question, then I can develop
my discussion, and try to take into account whatever
qualifications you elect to build into your answer. When
you avoid answering, that shuts off the development.
This is all what I have experienced several times, and I
have seen you do it with other correspondents. That is
what led me to make my previous comments about the futility
of trying to have "dialog" with you.
RF >
> If it's the FORMER, I'll play along as being the victim of your
> TROLL, for the reason that each time you do so, you lose just
> a bit more of whatever little credibility you might have had left
> in these sci.stat groups about your statistical knowledge. But
> I want to make CERTAIN that you are not misleading some
> innocent readers who are new to statistics into following your
> ERRORS, your statistical Quackery, and malpractice in
> statistics.
Bob, you have objected to me on, what, 5 issues in 1000
original threads? That is not very many. You've been alone.
And I have objected to you about as often,
usually in support of someone else.
That's not counting the ones generated by your attacking style.
>
> If it's the LATTER, may I be the first to congratulate you that
> you have given adequate PROOF that you are "infinitely
> obtuse". The proof was rather lengthy -- even more so than
> the 200-pages it took Wiles to prove Fermat's Last Theorem,
> but let me assure you that your proof is complete. Q.E.D.
>
> Was your post a TROLL or a PROOF?
>
> I'll let the readers decide.
I wish you would listen to readers. Here, and elsewhere.
They have frequently (for a group) asked you to "stop
the extraneous criticism of Rich Ulrich. It makes reading
unpleasant". Right now, I'm thinking of the last few days,
and the noise you keep injecting into every sci.stat.math
thread that I'm not in.
Reef Fish
Richard Ulrich wrote:
> On 21 Jul 2006 09:37:05 -0700, "Reef Fish"
> <Large_Nass...@yahoo.com> wrote:
> > >
> RU > >
> > > You concede that the mode belongs to a "family of
> > > means" that is defined by minimization?
> RF >
> > I conceded no such! The mode does NOT belong to any "family
> > of means". What the mean minimizes is IRRELEVANT to the
> > definition and meaning of the statistical MODE.
> >
> RU >
> > > But you think
> > > another "family of means" is not relevant?
> RF >
> > The "family of means" is IRRELEVANT, to the definition and
> > proper usage of the MODE. Period.
>
> Bob seems to have an odd attitude about classes and members,
> and what is apt to be defined, that is inappropriate. Or, he
> will not say why mention of the "family" is "irrelevant", when
> I think it is, and (I'm pretty sure) other people think so, too.
Now you put yourself as the spokesman for "other people" in this
group too?
> RF >
> > The "family of means" is IRRELEVANT, to the definition and
> > proper usage of the MODE. Period.
That is a straightfoward sentence in English. Why does Richard
has problem understanding it?
The MODE is the "most frequent" value -- which is NOT a mean
of any kind. So, how is any "family of means" relevant to it?
I think RIchard is using ABSURD arguments for his TROLL now.
> Consider one other idea/concept -- I looked up Human with
> the Google click (which brings up multiple reference books
> by using Answer.com). "Human" never mentions Mammal.
> The citations for Mammal more often do mention humans.
Did you want to find out the definition of "Human" or "Mammal"?
You are still wallowing in your own INABILITY to use the
English Dictionary properly, to look up the word for which
the definition is wanted: MODE.
Instead, Richard Ulrich is writing a book on how many ways he
can MIS-USE dictionary, encyclopedia, and now Google
information to get the WRONG definition and meaning of
a statistical term!
>
> For whatever his reason, Bob has been willing to accept the
> family of Generalized means as being real averages,
Because they ARE. But the MODE is none of them!
Richard, I had told you that you had already successfully proved
that you are "infinitely obtuse".
I suppose you just choose to continue your TROLL now, isn't it?
Well, each time you do it, you are putting one more nail into your
own coffin of ignorance and lack of credibility.
> RU >
> > > Do you insist on not-calling that a family of means
> > > (because it doesn't have to divide by N)?
> > > Or you don't call its members, "means"?
> RF >
> > I could care less what you call it. We were supposed to be
> > discussing what a MODE is. And the definition and meaning
> > and usage of the term MODE has nothing to do with what
> > any mean, or any member in the family of means.
Richard Ulrich's totally irrational and illogical obfuscation snipped.
>
> [...]
> RF >
> > Rich, at this point, it is my opinion, based on your latest posts
> > about the MODE, that you have entered the realm of being
> > a statistical TROLL, or you intend to prove that you're not only
> > obtuse in your understanding of statistics and statistical terms,
> > but you are INFINITELY obtuse.
>
> Bob, I keep trying to reduce my arguments to tinier steps,
> and stating the assumptions that I assumed were shared,
> in hopes that you can grasp them. When you give me
> no-comment-whatsoever on them, I don't have much to
> work with for articulating further.
I retained ALL my comments above, that had already fully
explained why Ulrich's arguments are totally invalid and
irrelevant. There is no more to add.
> RF >
> > If it's the FORMER, I'll play along as being the victim of your
> > TROLL, for the reason that each time you do so, you lose just
> > a bit more of whatever little credibility you might have had left
> > in these sci.stat groups about your statistical knowledge. But
> > I want to make CERTAIN that you are not misleading some
> > innocent readers who are new to statistics into following your
> > ERRORS, your statistical Quackery, and malpractice in
> > statistics.
> >
> > If it's the LATTER, may I be the first to congratulate you that
> > you have given adequate PROOF that you are "infinitely
> > obtuse". The proof was rather lengthy -- even more so than
> > the 200-pages it took Wiles to prove Fermat's Last Theorem,
> > but let me assure you that your proof is complete. Q.E.D.
> >
> > Was your post a TROLL or a PROOF?
> >
> > I'll let the readers decide.
>
> I wish you would listen to readers.
Did I miss someone recently who supported Ulrich's argument
that a MODE is a mean?
Let that reader speak up! I think we are all sick and tired of
Richard Ulrich's absurd argument that "the value of the
HIGHEST frequency" is a "mean" or "average" of a set of
statistical data.
None of the WORST of my freshman statistics students ever
made THAT bad a mistake.
Why is Richard Ulrich not only made the mistake, but continue
to argue that he has NOT made a mistake, over and over again?
He is a statistical Quack and now a TROLL.
-- Reef Fish Bob.
st...@mimosa.csv.warwick.ac.uk
Sum(w_i x_i) / Sum(w_i)
to be considered an "average"?
Making each w_i nonnegative seems sensible,
but in what ways can w_i involve the data?
For example, is a Winsorized mean
W_k: w_1=w_2=...=w_k = w_{n-k+1}=w_{n-k+2}=...=w_n = 0
w_{k+1} = w_{n-k} = k
w_{k+2}=w_{k+3}=...=w_{n-k-1} = 1
an average?
Is a grouped mean G (w_i is the frequency of x_i) an average?
Is G_k = Sum(w_i^k x_i) / Sum(w_i^k) an average?
Is G_infinity = lim_{k->infinity} G_k an average?
Is the arithmetico-geometric mean of two numbers an average?
...etc.
--
J.E.H.Shaw [Ewart Shaw] st...@uk.ac.warwick TEL: +44 2476 523069
Department of Statistics, University of Warwick, Coventry CV4 7AL, UK
http://www.warwick.ac.uk/statsdept http://www.ewartshaw.co.uk
3 ((4&({*.(=+/))++/=3:)@([:,/0&,^:(i.3)@|:"2^:2))&.>@]^:(i.@[) <#:3 6 2
Reef Fish
st...@mimosa.csv.warwick.ac.uk wrote:
> What restrictions should be placed on the weights w_i for
>
> Sum(w_i x_i) / Sum(w_i)
>
> to be considered an "average"?
Why does it matter?
As some of us on this side of the pond might say, "you're singing in
the
wrong choir and in the wrong church".
The subject is the definition and meaning of the word "MODE",
not pie a la mode or mode of transportation. It's the statistical
NOUN which is a summary statistic for a set of numbers. or a
description of thre relative maximum of a probability density function.
-- Reef Fish Bob.
Richard Ulrich
<Large_Nass...@yahoo.com> wrote:
>
> Richard Ulrich wrote:
> > On 21 Jul 2006 09:37:05 -0700, "Reef Fish"
> > <Large_Nass...@yahoo.com> wrote:
> > > >
> > RU > >
> > > > You concede that the mode belongs to a "family of
> > > > means" that is defined by minimization?
> > RF >
> > > I conceded no such! The mode does NOT belong to any "family
> > > of means". What the mean minimizes is IRRELEVANT to the
> > > definition and meaning of the statistical MODE.
> > >
> > RU >
> > > > But you think
> > > > another "family of means" is not relevant?
> > RF >
> > > The "family of means" is IRRELEVANT, to the definition and
> > > proper usage of the MODE. Period.
> >
> > Bob seems to have an odd attitude about classes and members,
> > and what is apt to be defined, that is inappropriate. Or, he
> > will not say why mention of the "family" is "irrelevant", when
> > I think it is, and (I'm pretty sure) other people think so, too.
>
> Now you put yourself as the spokesman for "other people" in this
> group too?
IMHO, the use of reasoning is forever appealing to the
agreement of others.
That post seems to mark the final words of Bob on the topic
of mode being a mean, so I will review.
In my first Reply, I stated -
The downside is that Bob is that I don't "dialog" with Bob,
because he can't. There is practically no chance that he
will have any relevant reply, and less chance that he will
admit he is wrong about either argument -- whether the mode
is a mean, and whether it matters much what someone asserts.
The upside is that Bob Ling has presented his argument.
I will show how little there is to it.
There were lines of definition for mode and mean, which Bob
selected from what was available in dictionaries. There were
ad-hominem comments intended for intimidation, against anyone
who opposed.
Bob never could or would try to defend his selection of
definitions. Many definitions of Average include explicit
naming of the mode; a few definitions of the Mode included
the mean. He is incapable of articulating why we should
prefer one set of definitions; repeating them is not articulation.
Or he is incapable of recognizing that "reasoning" requires
that he make a statement about the definitions.
His one 'argument' (or concession?) added in later posts is that
he endorsed as averages the members of one "family of means",
such as the geometric mean. That *seems* to say that Bob
abandons the simple common-sense definition of average
(raised by another poster) -- relevant, if he had wished to
try to stand on "simple common sense"
On Jul 23, Bob O'Hara quoted Bob, and asked pointedly,
RF >
> I had only posted at least 100 times that meaning of AVERAGE
> is IRRLEVANT to the defiition and meaning of the statistical
> MODE.
>
> If you want to find out the meaning of the MODE, you look up
> under the definition for the word MODE. That is the "FINAL
> ANSWER" in the four lines you cited.
>
Bob Anon >
"You're claiming that a mode is not an average, so surely we have to
know what an average is to be able to evaluate this claim. I could
define average in a way that's wide enough to include a mode as an
average, and then you would be wrong (according to my definition).
So, it's very relevant, and I'm surprised that you refuse."
Reef Fish Bob never replied.
*********** big summary ***********
The progress of the reasoning of Reef Fish Bob, therefore,
seems to be from
(a) offering a selection of definitions of average and mode, to
(b) reducing the selection to definitions of the mode.
*********** end summary ***********
That's less than convincing.
Jerry Dallal had posted some very wise words as early as
July 17. In a long post, I think he offered several versions
of this same essential point --
'I would never use "average" to describe a mode myself. However,
I am also not so blind as to fail to see that, outside the
profession, many people do it. Insofar as the vernacular goes, the
majority rules, for better or worse, like it or not. Which is also
not to say that I don't admire your quest. But, when things have
gotten to where mode/average is now, resistance is often futile, at
least outside the profession....'
That's a good statement. Bob has never responded to that one,
either.
I think I've said a couple of times that it would be nice if we had
a word that had one precise meaning, or a couple of words with
a couple of nice meanings. But we don't.
If Bob had qualified his proposition, "For statisticians
reporting ordinary data, the mode is never an average"...
I doubt that he would have had any argument in any Usenet
group.
Following his original offering-of-the-dictionaries, Bob
wrote more lines offering of ad-hominem-with-intimidation.
[snip, about 90 lines: verbal abuse, characterizing me,
TROLL ... dictionary use ... obtuse ... TROLL
irrational ... invalid and irrelevant ... TROLL 5 or six
times in all.]
The original intimidation is now refocussed -- It has
migrated to "Rich Ulrich is a troll." I've mentioned
before that "verbal reasoning" is the sort of thing that
Bob identifies as "irrelevant" and "irrational" and "nonsense"
and so on; that seems to apply here.
I quit thinking of Bob Ling as a troll when I realized that he
must be sincere with this mode/mean nonsense in particular.
That leaves him, to me, as "some variety of net.kook."
>
> Did I miss someone recently who supported Ulrich's argument
> that a MODE is a mean?
>
> Let that reader speak up!
Jerry spoke up. He said, the majority rules.
"Insofar as the vernacular goes, the majority rules,
for better or worse, like it or not." [cited above]
Bob spoke up. He said, surely the definition of Average
has to matter.
> I think we are all sick and tired of
I think we are "all sick and tired" of Bob saying that Rich
Ulrich is not a serious, intelligent respondent. -- and
plunking it into irrelevant threads. If you want to talk
about "sick and tired."
> Richard Ulrich's absurd argument that "the value of the
> HIGHEST frequency" is a "mean" or "average" of a set of
> statistical data.
Clearly, that's not the argument as I made it. That phrase
tacked on at the end is the grosser infraction.
So. Is Bob re-stating the proposition to something new?
- He should have tried that approach two years ago.
But I'm *pretty* sure that he has done that before, and
then hastened back to most general statement, claiming
they are the same thing.
Wording matters. "For statisticians" is rather like, "of a
set of statistical data." Both of them imply something
limiting. I'd probably go along with those.
If he wants to argue something other than the *most*
general case, as he insisted on stating it, perhaps he
should offer it as an apology to the bridge Group.
But I think he's spouting words from the top of his brain
again.
Here is what I quoted at the start of this Reply, from my
first Reply in the Mode-Average thread --
The downside is that Bob is that I don't "dialog" with Bob,
because he can't. There is practically no chance that he
will have any relevant reply, and less chance that he will
admit he is wrong about either argument -- whether the mode
is a mean, and whether it matters much what someone asserts.
That seems to have been a valid prediction, mainly.
One bit of 'responsiveness' -- he shortened his list of definitions,
so that *all* the definitions of Average should be discarded.
There was an interesting digression concerning the
"intimidation" after I responded that surely Bob's error
on Independence was "discrediting", if anything was.
Bob couldn't figure the logic of that but brought in
my defense of my jargon - something that turns out
to be Cohen's own jargon - for power analysis, which
does not always specify "rate."
If a few other teachers came in to support Jerry, I would
probably come to agree that the jargon should be avoided,
If Bob's reaction to Cohen was shown in cartoon form, I
suppose Bob's head explodes. In a couple of separate posts, he
abuses Cohen almost as harshly as he ordinarily abuses me.
Since Bob still has not read Cohen, and since Bob still knows
essentially nothing about the behavioral sciences, that seems
to be an ill-advised approach to posting. But it's probably,
once again, just something off the top of his brain.
>
> None of the WORST of my freshman statistics students ever
> made THAT bad a mistake.
IMHO, Bob is really poor on judgments and comparisons of
that sort - what's a mistake, or what is a *bad* mistake.
Too prone to exaggeration. No empathy. No respect for
other points of view.
Reef Fish
Richard Ulrich ha escrito:
> On 22 Jul 2006 20:15:04 -0700, "Reef Fish"
> <Large_Nass...@yahoo.com> wrote:
>
> > > RU > >
> > > > > You concede that the mode belongs to a "family of
> > > > > means" that is defined by minimization?
> > > RF >
> > > > I conceded no such! The mode does NOT belong to any "family
> > > > of means". What the mean minimizes is IRRELEVANT to the
> > > > definition and meaning of the statistical MODE.
> That post seems to mark the final words of Bob on the topic
> of mode being a mean, so I will review.
What a clever wording to IMPLY that I had agreed that a
mode is a mean, as he previously tried to imply the same
when I gave him a flat NO.
> > > > > You concede that the mode belongs to a "family of
> > > > > means" that is defined by minimization?
Then Richard Ulrich lauched into his diatribe about Bob not
reading well, and not writing well, etc., etc.
Richard, there can't be a simpler English sentence than;
A statistical MODE is NOT an Average (or Mean).
-- Reef Fish
Anon.
> Richard Ulrich ha escrito:
>
<snip>
>>>>>> You concede that the mode belongs to a "family of
>>>>>> means" that is defined by minimization?
>
> Then Richard Ulrich lauched into his diatribe about Bob not
> reading well, and not writing well, etc., etc.
>
> Richard, there can't be a simpler English sentence than;
>
> A statistical MODE is NOT an Average (or Mean).
>
it depends on how one understands the word "average"?
(the "(or Mean)" is a new addition, the discussion was about the
sentence "A statistical MODE is NOT an Average", so I hope we'll stick
to it: it's even clearer!)
Bob
--
Bob O'Hara
Department of Mathematics and Statistics
P.O. Box 68 (Gustaf Hällströmin katu 2b)
FIN-00014 University of Helsinki
Finland
Telephone: +358-9-191 51479
Mobile: +358 50 599 0540
Fax: +358-9-191 51400
WWW: http://www.RNI.Helsinki.FI/~boh/
Journal of Negative Results - EEB: www.jnr-eeb.org
Reef Fish
Anon. ha escrito:
> Reef Fish wrote:
> > Richard Ulrich ha escrito:
> >
> <snip>
> >>>>>> You concede that the mode belongs to a "family of
> >>>>>> means" that is defined by minimization?
> >
> > Then Richard Ulrich lauched into his diatribe about Bob not
> > reading well, and not writing well, etc., etc.
> >
> > Richard, there can't be a simpler English sentence than;
> >
> > A statistical MODE is NOT an Average (or Mean).
> >
> Indeed. So isn't it clear that a decision about whether one agrees with
> it depends on how one understands the word "average"?
>
> (the "(or Mean)" is a new addition, the discussion was about the
> sentence "A statistical MODE is NOT an Average", so I hope we'll stick
> to it: it's even clearer!)
Bob O'Hara, you have missed nearly ALL of the discussions, which
explained your clueless remarks. It's rather obvious you did not go
back and try to read some of the earlier posts.
>From the very outset, NO ONE did not understand Average to mean
the arithmetic average (or any other kind of MEAN).
>From the very outset, NO ONE misunderstands what a MEAN is.
>From the very outset, the ones who are IGNORANT, and insist to
be IGNORANT about the meaning of a MODE are those who
refuse to take the DEFINITION of a MODE (the highest frequency)
as the statistical meaning and immediately see that it is NOT a
MEAN or an AVERAGE.
It's THAT simple.
The word AVERAGE in the English Dictionary is nothing but a
red-herring among the uneducated in arguing against
A statistical MODE is NOT an Average (or Mean, or any class
of means or anything that can be construed to be an AVERAGE).
A MODE has a stand-alone meaning in statistics. Period!
-- Reef Fish
Anon.
> Anon. ha escrito:
>
>> Reef Fish wrote:
>>> Richard Ulrich ha escrito:
>>>
>> <snip>
>>>>>>>> You concede that the mode belongs to a "family of
>>>>>>>> means" that is defined by minimization?
>>> Then Richard Ulrich lauched into his diatribe about Bob not
>>> reading well, and not writing well, etc., etc.
>>>
>>> Richard, there can't be a simpler English sentence than;
>>>
>>> A statistical MODE is NOT an Average (or Mean).
>>>
>> Indeed. So isn't it clear that a decision about whether one agrees with
>> it depends on how one understands the word "average"?
>>
>> (the "(or Mean)" is a new addition, the discussion was about the
>> sentence "A statistical MODE is NOT an Average", so I hope we'll stick
>> to it: it's even clearer!)
>
> Bob O'Hara, you have missed nearly ALL of the discussions, which
> explained your clueless remarks. It's rather obvious you did not go
> back and try to read some of the earlier posts.
>
clarify this point: I have a suspicion that problems over deciding
whether a mode is an average derive from different ideas about what
"average" means.
>>From the very outset, NO ONE did not understand Average to mean
> the arithmetic average (or any other kind of MEAN).
>
>>From the very outset, NO ONE misunderstands what a MEAN is.
>
>
>>From the very outset, the ones who are IGNORANT, and insist to
> be IGNORANT about the meaning of a MODE are those who
> refuse to take the DEFINITION of a MODE (the highest frequency)
> as the statistical meaning and immediately see that it is NOT a
> MEAN or an AVERAGE.
>
> It's THAT simple.
>
> The word AVERAGE in the English Dictionary is nothing but a
> red-herring among the uneducated in arguing against
>
So knowing what "average" means is irrelevant to knowing whether
something is an average?
> A statistical MODE is NOT an Average (or Mean, or any class
> of means or anything that can be construed to be an AVERAGE).
>
> A MODE has a stand-alone meaning in statistics. Period!
>
So does "arithmetic mean". Does that mean that's not an average either?
I'm hoping for some enlightenment: that you might explain your
reasoning. I'd like to understand your point of view, but it's
difficult when all you do is make statements and refuse to back them up.
Reef Fish
Anon. ha escrito:
> Reef Fish wrote:
> > Anon. ha escrito:
> >>> Richard, there can't be a simpler English sentence than;
> >>>
> >>> A statistical MODE is NOT an Average (or Mean).
> >>>
> >> Indeed. So isn't it clear that a decision about whether one agrees with
> >> it depends on how one understands the word "average"?
> > Bob O'Hara, you have missed nearly ALL of the discussions, which
> > explained your clueless remarks. It's rather obvious you did not go
> > back and try to read some of the earlier posts.
> >
> I have been following the discussions, which is why I'm trying to
> clarify this point: I have a suspicion that problems over deciding
> whether a mode is an average derive from different ideas about what
> "average" means.
Only in the sense that those who erred are in ERROR on BOTH the
meaning of AVERAGE and definition of the MODE.
> >>From the very outset, NO ONE did not understand Average to mean
> > the arithmetic average (or any other kind of MEAN).
> >
> >>From the very outset, NO ONE misunderstands what a MEAN is.
> >
> >
> >>From the very outset, the ones who are IGNORANT, and insist to
> > be IGNORANT about the meaning of a MODE are those who
> > refuse to take the DEFINITION of a MODE (the highest frequency)
> > as the statistical meaning and immediately see that it is NOT a
> > MEAN or an AVERAGE.
> >
> > It's THAT simple.
> >
> > The word AVERAGE in the English Dictionary is nothing but a
> > red-herring among the uneducated in arguing against
> So knowing what "average" means is irrelevant to knowing whether
> something is an average?
It is relevant to know that the mode is NOT any kind of average.
It is an ERROR to think that any AVERAGE is a MODE, notwithstanding
many English Dictictionary did make that ERROR.
But if one knows how to use an English dictionary (see the SUBJECT)
then one would look for the definition of the MODE, and in every single
case, it would have given the correct definition. Richard Ulrich was
challenged to find an English dictionary on the word MODE that does
not define it correctly, as the "most frequent" value, which of course
even a 4-year-old would know is not an average.
Richard Ulrich was not able to find an English Dictionary that
explains the statistical MODE as an average, any average.
Bob O'Hara, perhaps your English is part of your problem besides
NOT having read the lengthy thread (even though you said you
tried to follow).
This will be my FINAL reply to your questions -- I had already bent
backwards to accommodate your NOT having read what went on
before.
If you still don't understand, just go back and RE-READ the
explanations
until you do. (I am making the good-faith assumption that you
are not just playing dumb).
>
> > A statistical MODE is NOT an Average (or Mean, or any class
> > of means or anything that can be construed to be an AVERAGE).
> >
> > A MODE has a stand-alone meaning in statistics. Period!
READ THE ABOVE SUMMARY AGAIN AND AGAIN.
Then you wouldn't be asking the stupid question below;
> >
> So does "arithmetic mean". Does that mean that's not an average either?
>
> I'm hoping for some enlightenment: that you might explain your
> reasoning. I'd like to understand your point of view, but it's
> difficult when all you do is make statements and refuse to back them up.
I backed it up a hundred times with the DEFINITION in Statistics.
I backed it up a dozen or more times with the English Dictionary (see
the first post of this thread), as long as one looks up the English
dictionary definition for the word MODE, and not some other word.
In every single case, it is unequivocal that the MODE is NOT an
Average.
-- Reef Fish Bob.
Richard Ulrich
<Large_Nass...@yahoo.com> wrote:
[snip]
>
> Richard Ulrich was not able to find an English Dictionary that
> explains the statistical MODE as an average, any average.
>
Astonishment! Bob misses more posts.
I had given Bob two definitions *before* he defied me to find one.
Unfortunately, Bob was unfamiliar with accessing multiple
dictionaries at Answer.com from using the Google shortcut,
so he ignored the answer, which he was quoting from.
I repeated the information with detail, after he posed the
'challenge' -- he ignored or missed that, too. Apparently.
===== 35 lines of what I posted on July 19, in Reply to July 18.
RU > > - As to the challenge -
> >
> > Columbia University Press -
> > mode, in statistics, an infrequently used type of average. In a
> > group of numbers the mode is the number occurring most frequently.
> > In the group 1, 4, 5, 5, 6, 6, 6, 6, 9, 9, the mode is 6 because it
> > occurs four times and the others only once or twice.
> >
> > Oops! That says "encyclopedia." Close enough?
> > Close enough for me. See the others, too.
RF >
> So, after extensive search, Richard Ulrich cannot find ANY dictionary
> that would give a wrong definition of the mode.
One click, the first place I looked, is hardly an extensive search.
Comments.
1) As references go, why is an encyclopedia inadequate?
2) I figured that I deserved some extra credit for having
answered the question for Bob, a day earlier than the question.
Even though he could not read it. Or, did he see the encyclopedia
citation and want a separate one?
3) Two days ago, another user cited and quoted a dictionary,
which was a statistical dictionary.
Now, what was the point of the Challenge?
Was there an implication that Bob would admit ignominious
defeat?
Naah...
====== end of July 19.
So Bob had two citations before he asked. Is there any
reason to fault an encyclopedia, except that its listing
may be more complete?
Here is a third dictionary reference --
==== start of cite from http://www.bartleby.com/59/19/mode.html
The New Dictionary of Cultural Literacy, Third Edition. 2002.
mode
In statistics, the most frequently appearing value in a set of
numbers or data points. In the numbers 1, 2, 4, 6, 8, 4, 9, 6, 8,
and 6, the mode is 6, because it appears more often than any of the
other figures. (See average; compare mean and median.) 1
==== end of cite ... see "referred cite" a few lines down.
Here is another note on How to use a dictionary: When it says,
"See average", the dictionary is telling where there is more
information. Here is the whole citation at "average" --
==== "referred cite"
average
A single number that represents a set of numbers. Means, medians,
and modes are kinds of averages; usually, however, the term average
refers to a mean.
==== end of cite
Once again, this says modes "are kinds of averages."
I mentioned "usage notes" some time ago, probably by
that name -- This fits under the present Subject: of
dictionary usage. Where there has been some doubt,
it is not uncommon to see some explanation like this.
Following SEE from mode, the wiki-dictionary
Wiktionary has this "usage note."
This is a fourth reference, 3rd dictionary, that says directly or
posts pointers to say that the mode is an average.
http://en.wiktionary.org/wiki/average
==== start of cite
Usage notes
'The term average may refer to the statistical mean, median or mode
of a distribution, or sometimes any other measure of central
tendency. Statisticians and responsible news sources are careful to
use whichever of these specific terms is appropriate. In common
usage, average refers to the arithmetic mean. It is, however, a
common rhetorical trick to call the most favorable of mean, median
and mode the "average" depending on the interpretation of a set of
figures that the speaker or writer wants to promote.'
==== end of cite.
The first line says what matters for Bob's "challenge", but the
rest is interesting, too.
[snip, more of Bob's post]
> I backed it up a hundred times with the DEFINITION in Statistics.
> I backed it up a dozen or more times with the English Dictionary (see
> the first post of this thread), as long as one looks up the English
> dictionary definition for the word MODE, and not some other word.
>
> In every single case, it is unequivocal that the MODE is NOT an
> Average.
Bob will decide whether A is a subset of B, or a
subspecies of B, independent of the constitution of B.
What school of logic teaches that?
Ipso-facto, Bob's contention demonstrates his inability
to use formal logic and verbal reasoning as the rest of
the world does.
That violation of basic reasoning seems like a red flag,
"Hey, I'm not competent to discuss anything at all" --
Half a dozen idiosyncratic usages of words seem to me
like nothing, compared to the failure of logic itself.
--
Rich Ulrich, wpi...@Pitt.edu
http://www.pitt.edu/~wpilib/index.html
Anon.
> Anon. ha escrito:
>
<snip>
> Then you wouldn't be asking the stupid question below;
>> So does "arithmetic mean". Does that mean that's not an average either?
>>
>> I'm hoping for some enlightenment: that you might explain your
>> reasoning. I'd like to understand your point of view, but it's
>> difficult when all you do is make statements and refuse to back them up.
>
> I backed it up a hundred times with the DEFINITION in Statistics.
Of "mode". Not of "average".
Are you genuinely having finding it impossible to understand my point:
that we need to understand what "average" means? I've never asked for a
definition of "mode".
st...@mimosa.csv.warwick.ac.uk
"Reef Fish" <Large_Nass...@yahoo.com> writes:
>
>st...@mimosa.csv.warwick.ac.uk wrote:
>> What restrictions should be placed on the weights w_i for
>>
>> Sum(w_i x_i) / Sum(w_i)
>>
>> to be considered an "average"?
>
>Why does it matter?
One reason it matters is that the mode (of a unimodal data set)
is the limit of just such a weighted average - see the part of
my post that you snipped. I still want to know where the boundary is
between what you consider to be an average, and what you don't.
When you say refer to an "average", do you allow anything other than
the arithmetic mean, and if so, what?
-- Many thanks, Ewart Shaw
>
>As some of us on this side of the pond might say, "you're singing in
>the
>wrong choir and in the wrong church".
>
>The subject is the definition and meaning of the word "MODE",
>not pie a la mode or mode of transportation. It's the statistical
>NOUN which is a summary statistic for a set of numbers. or a
>description of thre relative maximum of a probability density function.
>
>-- Reef Fish Bob.
Reef Fish
st...@mimosa.csv.warwick.ac.uk ha escrito:
> In article <1154043823.7...@i3g2000cwc.googlegroups.com>,
> "Reef Fish" <Large_Nass...@yahoo.com> writes:
> >
> >st...@mimosa.csv.warwick.ac.uk wrote:
> >> What restrictions should be placed on the weights w_i for
> >>
> >> Sum(w_i x_i) / Sum(w_i)
> >>
> >> to be considered an "average"?
You were ignored because it is IRRELEVANT to the meaning of the MODE.
> >
> >Why does it matter?
>
> One reason it matters is that the mode (of a unimodal data set)
> is the limit of just such a weighted average - see the part of
> my post that you snipped.
What limit? For a unimodal data set, it is the value of the HIGHEST
FREQUENCY.
I still want to know where the boundary is
> between what you consider to be an average, and what you don't.
Whatever you consider to be an average, the statistical MODE is NOT
that defined to be any average.
> When you say refer to an "average", do you allow anything other than
> the arithmetic mean, and if so, what?
> -- Many thanks, Ewart Shaw
You should have reviewed what had gone long before you jumped in.
It allowed the geometric mean, harmonic mean, trimmed mean,
winsorized mean, etc., etc. Any mean or average you care to use,
the MODE is NOT one of them.
>
> >
> >As some of us on this side of the pond might say, "you're singing in
> >the
> >wrong choir and in the wrong church".
> >
> >The subject is the definition and meaning of the word "MODE",
> >not pie a la mode or mode of transportation. It's the statistical
> >NOUN which is a summary statistic for a set of numbers. or a
> >description of thre relative maximum of a probability density function.
> >
> >-- Reef Fish Bob.
This being the end of another month -- I hope this tedious thread
about the SIMPLEST statistical definition and notion of a MODE
(that of a point of the HIGHEST FREQUENCY or a point of relative
maximum in a continuous density) will end in a STATISTICAL group,
after it has been ongoing for 15 MONTHS.
Those who are still muddled in this simplest of the statistical
concepts
should take up another field or discipline OTHER than statistics.
-- Reef Fish Bob.
-- Reef Fish Bob.
Anon.
> st...@mimosa.csv.warwick.ac.uk ha escrito:
>
<snip>
>> When you say refer to an "average", do you allow anything other than
>> the arithmetic mean, and if so, what?
>> -- Many thanks, Ewart Shaw
>
> You should have reviewed what had gone long before you jumped in.
> It allowed the geometric mean, harmonic mean, trimmed mean,
> winsorized mean, etc., etc. Any mean or average you care to use,
> the MODE is NOT one of them.
Does the median count as an average?
>>> As some of us on this side of the pond might say, "you're singing in
>>> the
>>> wrong choir and in the wrong church".
>>>
>>> The subject is the definition and meaning of the word "MODE",
>>> not pie a la mode or mode of transportation. It's the statistical
>>> NOUN which is a summary statistic for a set of numbers. or a
>>> description of thre relative maximum of a probability density function.
>>>
>>> -- Reef Fish Bob.
>
> This being the end of another month -- I hope this tedious thread
> about the SIMPLEST statistical definition and notion of a MODE
> (that of a point of the HIGHEST FREQUENCY or a point of relative
> maximum in a continuous density) will end in a STATISTICAL group,
> after it has been ongoing for 15 MONTHS.
>
The problem is not the notion of a mode, but the notion of an average.
Reef Fish
Anon. wrote:
> Reef Fish wrote:
> > st...@mimosa.csv.warwick.ac.uk ha escrito:
> >
> <snip>
> >> When you say refer to an "average", do you allow anything other than
> >> the arithmetic mean, and if so, what?
> >> -- Many thanks, Ewart Shaw
> >
> > You should have reviewed what had gone long before you jumped in.
> > It allowed the geometric mean, harmonic mean, trimmed mean,
> > winsorized mean, etc., etc. Any mean or average you care to use,
> > the MODE is NOT one of them.
>
> Does the median count as an average?
NO.
>
> >>> As some of us on this side of the pond might say, "you're singing in
> >>> the
> >>> wrong choir and in the wrong church".
> >>>
> >>> The subject is the definition and meaning of the word "MODE",
> >>> not pie a la mode or mode of transportation. It's the statistical
> >>> NOUN which is a summary statistic for a set of numbers. or a
> >>> description of thre relative maximum of a probability density function.
> >>>
> >>> -- Reef Fish Bob.
> >
> > This being the end of another month -- I hope this tedious thread
> > about the SIMPLEST statistical definition and notion of a MODE
> > (that of a point of the HIGHEST FREQUENCY or a point of relative
> > maximum in a continuous density) will end in a STATISTICAL group,
> > after it has been ongoing for 15 MONTHS.
> >
> The problem is not the notion of a mode, but the notion of an average.
Of course. The English word "average" is NOT a statistical term.
It is undefined and ill-defined. But whatever you want to call an
"average"
it is NOT a mode, and it does NOT include a MODE.
To get yourself out fo the same quagmire that sank Ulrich and other
educated in statistics, you at least should know enough mathematics
to see the various examples of the BETA distribution that a MODE is
NOT an average, for the U-shaped, and J-shaped distributions, and
I don't care how you have misused the English or other dictionaries
for the word "average".
You should be able to see clearly what the MODE is and WHY it is
not an average.
If not, perhaps Luis Afonso and OMU will give Richard Ulrich and you
a joint clinic on the meaning of the statistical MODE.
> --
> Bob O'Hara
> Department of Mathematics and Statistics
And you don't know what a MODE is????
Shocking beyond belief. I thought your failure to understand what
a "linear regression model" is was shocking enough, after weeks and
weeks of lectures on something I normally cover in half a class. But
you clearly topped yourself this time.
-- Reef Fish Bob.
Richard Ulrich
this morning. I haven't seen his usual instant-commentary. -
On 31 Jul 2006 15:32:14 -0700, "Reef Fish"
<Large_Nass...@yahoo.com> wrote:
>
> Anon. wrote:
> > Reef Fish wrote:
> > > st...@mimosa.csv.warwick.ac.uk ha escrito:
[snip]
>
> To get yourself out fo the same quagmire that sank Ulrich and other
> educated in statistics, you at least should know enough mathematics
> to see the various examples of the BETA distribution that a MODE is
> NOT an average, for the U-shaped, and J-shaped distributions, and
> I don't care how you have misused the English or other dictionaries
> for the word "average".
Any form of averages is only good part of the time.
I've run into more distributions that had negative numbers
than the ones that are U-shaped. The geometric mean is not
an average for negatives numbers, but I've learned to use
"whatever is appropriate" (or useful). The mode can be useful.
The arithmetic mean performs pretty uselessly whenever there
are great outliers -- that suggests why we use a variety of
"averages", in order to represent what is central or typical.
The mode can be more central and typical, for some particular
data.
- What I just did above is called "reasoning". I generalized what
Bob said, and applied it elsewhere, to negate his own point.
It is also called "articulating" since it looks at related cases of
the particular that Bob drew upon. Bob seldom argues this way,
although it is true that he is "articulating" this "Mode" argument
more than he has done before, when he mentions the examples
of Beta and other distributions.
>
> You should be able to see clearly what the MODE is and WHY it is
> not an average.
I did not have a firm opinion on whether the "MODE" should
be considered an "average", or when it should, until *after*
I read up on Bob's arguments. Bob's "arguments" repelled.
I saw how badly he was stating both the problem and the arguments.
He was flopping down a proposition in the most indiscriminate way,
and calling the world STUPID for disagreeing with his insistence.
He has never thought of making an effort to construct an
intelligent statement about Modes and means -- one that everyone
might agree on.
I think that the Usage Note I quoted in my post from 3:00 this
morning is very nice. It makes some of Bob's point - the "mode"
may be a bit disreputable. However, it does concede that the mode
is an average, which is the only *reasoning* conclusion that
is possible, for the formal proposition as he stated it. (Lately, he
keeps trying to restate it differently.... that could be a start for
a discussion, if he knew how to discuss.)
Bob can accept whatever beliefs he wants on faith as being
"proper", but his "religious preference" is not a stable ground
for disparaging people who believe differently. Or who
point out that he is not using reason (since, mostly, he isn't).
[snip ...]
Reef Fish
Richard Ulrich wrote:
> - I hope Reef Fish Bob did not overlook my long post of early
> this morning. I haven't seen his usual instant-commentary. -
>
Mexico back to the US.
You are pitifully STUPID.
My reply to Anon Bob O'Hara and Richard Ulrich's noise snipped.
I never thought I would see the day that Luis A. Afonso said of
YOU Richard Ulrich --
Afonso> UNMISTAKABLY (a)
Afonso> __ MODE is NOT an average.
Afonso> I fell strongly that I AM (absolutely) UNABLE to switch
Afonso> the opinion of your opponents. Consequently I do not
Afonso> take part of the *discussion*. NEVER.
That put a BIG smile on my face!
That even Afonso sees immediately how obtuse and dense you are.
-- Reef Fish Bob.
Anon.
> Anon. wrote:
<snip>
>>>>> As some of us on this side of the pond might say, "you're singing in
>>>>> the
>>>>> wrong choir and in the wrong church".
>>>>>
>>>>> The subject is the definition and meaning of the word "MODE",
>>>>> not pie a la mode or mode of transportation. It's the statistical
>>>>> NOUN which is a summary statistic for a set of numbers. or a
>>>>> description of thre relative maximum of a probability density function.
>>>>>
>>>>> -- Reef Fish Bob.
>>> This being the end of another month -- I hope this tedious thread
>>> about the SIMPLEST statistical definition and notion of a MODE
>>> (that of a point of the HIGHEST FREQUENCY or a point of relative
>>> maximum in a continuous density) will end in a STATISTICAL group,
>>> after it has been ongoing for 15 MONTHS.
>>>
>> The problem is not the notion of a mode, but the notion of an average.
>
> Of course. The English word "average" is NOT a statistical term.
> It is undefined and ill-defined. But whatever you want to call an
> "average"
> it is NOT a mode, and it does NOT include a MODE.
>
some room for uncertainty about whether a particular statistic is an
average or not. That's why I want to make sure we're clear on what you
mean by "average".
> To get yourself out fo the same quagmire that sank Ulrich and other
> educated in statistics, you at least should know enough mathematics
> to see the various examples of the BETA distribution that a MODE is
> NOT an average, for the U-shaped, and J-shaped distributions, and
> I don't care how you have misused the English or other dictionaries
> for the word "average".
>
> You should be able to see clearly what the MODE is and WHY it is
> not an average.
>
For bimodal beta distributions, I would suggest that the modes can give
a decent summary of the location of the probability mass: near 0 and 1.
In that sense, they are summaries of location, which could be a
reasonable definition of an average.
> If not, perhaps Luis Afonso and OMU will give Richard Ulrich and you
> a joint clinic on the meaning of the statistical MODE.
>
>> --
>> Bob O'Hara
>> Department of Mathematics and Statistics
>
> And you don't know what a MODE is????
Oh, for ****s sake! Read through my posts again. You'll see that I've
repeatedly stated that "mode" isn't the problem., it's "average". I
thought I had finally got this through to you, but perhaps not.
Bob
--
Bob O'Hara
Department of Mathematics and Statistics
Reef Fish
That is your error in LOGIC. A mode is very well defined.
Luis A. Afonso has refused to explain to you and Richard Ulrich the
simple fact and definition of a MODE:
Afonso> UNMISTAKABLY (a)
Afonso> __ MODE is NOT an average.
Afonso> Because the definitions of each one of them.
> > To get yourself out fo the same quagmire that sank Ulrich and other
> > educated in statistics, you at least should know enough mathematics
> > to see the various examples of the BETA distribution that a MODE is
> > NOT an average, for the U-shaped, and J-shaped distributions, and
> > I don't care how you have misused the English or other dictionaries
> > for the word "average".
> >
> > You should be able to see clearly what the MODE is and WHY it is
> > not an average.
> >
> For bimodal beta distributions, I would suggest that the modes can give
> a decent summary of the location of the probability mass: near 0 and 1.
> In that sense, they are summaries of location, which could be a
> reasonable definition of an average.
You are just sinking DEEPER into your abyss of ignorance.
>
> > If not, perhaps Luis Afonso and OMU will give Richard Ulrich and you
> > a joint clinic on the meaning of the statistical MODE.
Luis Afonso has already refused. (See above. It was obvious to him).
We'll have to see of Old Mac User or anyone else wants to take part
in the Theatre of the Absurd starring Ulrich and O'Hara.
-- Reef Fish Bob.
Anon.
>>> Of course. The English word "average" is NOT a statistical term.
>>> It is undefined and ill-defined.
(source: 14 lines up)
>>> To get yourself out fo the same quagmire that sank Ulrich and other
>>> educated in statistics, you at least should know enough mathematics
>>> to see the various examples of the BETA distribution that a MODE is
>>> NOT an average, for the U-shaped, and J-shaped distributions, and
>>> I don't care how you have misused the English or other dictionaries
>>> for the word "average".
>>>
>>> You should be able to see clearly what the MODE is and WHY it is
>>> not an average.
>>>
>> For bimodal beta distributions, I would suggest that the modes can give
>> a decent summary of the location of the probability mass: near 0 and 1.
>> In that sense, they are summaries of location, which could be a
>> reasonable definition of an average.
>
> You are just sinking DEEPER into your abyss of ignorance.
>
Ah, no explanation of why I'm wrong, only insult. Please can you
explain why my argument is wrong.
Reef Fish
Richard Ulrich wrote:
> [snip]
> >
> > To get yourself out fo the same quagmire that sank Ulrich and other
> > educated in statistics, you at least should know enough mathematics
> > to see the various examples of the BETA distribution that a MODE is
> > NOT an average, for the U-shaped, and J-shaped distributions, and
> > I don't care how you have misused the English or other dictionaries
> > for the word "average".
The beta distribution is a DENSITY function whose prerequisite
is that f(x) .GE. 0.
>
> I've run into more distributions that had negative numbers
> than the ones that are U-shaped.
Negative numbers for a density function? Or negative values
or a r.v. ? Both are perfectly impertinent to the beta distribution
in question AND their modes.
-- Reef Fish Bob.
Reef Fish
If a MODE is well defined and "average" is undefined and ill-defined.
how can something well-defined by something ill-defined? That's
why I said
> > That is your error in LOGIC. A mode is very well defined.
>
> >>> To get yourself out fo the same quagmire that sank Ulrich and other
> >>> educated in statistics, you at least should know enough mathematics
> >>> to see the various examples of the BETA distribution that a MODE is
> >>> NOT an average, for the U-shaped, and J-shaped distributions, and
> >>> I don't care how you have misused the English or other dictionaries
> >>> for the word "average".
> >>>
> >>> You should be able to see clearly what the MODE is and WHY it is
> >>> not an average.
> >>>
> >> For bimodal beta distributions, I would suggest that the modes can give
> >> a decent summary of the location of the probability mass: near 0 and 1.
> >> In that sense, they are summaries of location, which could be a
> >> reasonable definition of an average.
> >
> > You are just sinking DEEPER into your abyss of ignorance.
> >
> Ah, no explanation of why I'm wrong, only insult. Please can you
> explain why my argument is wrong.
Anon Bob O'Hara, it is IMPOSSIBLE to say anything that can be
construed to be an insult to you. You are so utterly STUPID and
DENSE that to call you an MORON or an IMBECILE would only
be a compliment of your IQ to a point you can never attain.
You have proven your inability to reason or understand the simplest
of all statistical concepts and terms in your presence in
sci.stat.math,
that it's beyond words I can use to see how someone like you can
be in a department of statistics -- presumably doing SOME
work in statistics and not just a janitor.
Perhaps you ARE a Janitor in your department? If so, that would
explain your inability to learn statistics.
-- Reef Fish Bob.
st...@mimosa.csv.warwick.ac.uk
"Reef Fish" <Large_Nass...@yahoo.com> writes:
>
>st...@mimosa.csv.warwick.ac.uk ha escrito:
>
>> In article <1154043823.7...@i3g2000cwc.googlegroups.com>,
>> "Reef Fish" <Large_Nass...@yahoo.com> writes:
>> >
>> >st...@mimosa.csv.warwick.ac.uk wrote:
>> >> What restrictions should be placed on the weights w_i for
>> >>
>> >> Sum(w_i x_i) / Sum(w_i)
>> >>
>> >> to be considered an "average"?
>
>You were ignored because it is IRRELEVANT to the meaning of the MODE.
>> >
>> >Why does it matter?
>>
>> One reason it matters is that the mode (of a unimodal data set)
>> is the limit of just such a weighted average - see the part of
>> my post that you snipped.
>
>What limit? For a unimodal data set, it is the value of the HIGHEST
>FREQUENCY.
There was only one limit mentioned in my post. To reiterate:
I assume that the grouped mean G, where w_i = f_i (frequency of x_i),
is uncontroversially an average. If you fix k, then w_i = f_i^k
also gives an average G_k. If the dataset is unimodal, then
lim_{k->infinity) G_k is the mode. There's a case for including
this limit as an "average", in the same way as log is included
in the family of "power transformations".
If no restrictions are put on w_i, then you could of course
obtain the mode directly (in the unimodal case) by
w_i = 1 if f_i = max_j(f_j),
w_i = 0 otherwise.
Since you allow e.g. Winsorized mean, for which w_i uses knowledge
additional to the corresponding x_i, I want to know precisely what
additional knowledge you consider allowable.
As another example, if you want to consider increasing sample size,
then w_i = f_i^f_i gives an average M that converges almost surely
to the mode. Even when sampling from a beta distribution with
alpha = beta = 1/2, which you mentioned in another post, M has
asymptotic distribution with probability 1/2 at 0 and at 1.
[Assuming observations can only be recorded to finite precision:-)]
>I still want to know where the boundary is
>> between what you consider to be an average, and what you don't.
>
>Whatever you consider to be an average, the statistical MODE is NOT
>that defined to be any average.
>
>> When you say refer to an "average", do you allow anything other than
>> the arithmetic mean, and if so, what?
>> -- Many thanks, Ewart Shaw
>
>You should have reviewed what had gone long before you jumped in.
>It allowed the geometric mean, harmonic mean, trimmed mean,
>winsorized mean, etc., etc. Any mean or average you care to use,
>the MODE is NOT one of them.
You allow the Winsorized mean, but in another post you emphatically
said ("NO") that the median is not an average. But for sample size
2k+1 or 2k+2, the k-times Winsorized mean is just the median.
This illustrates why it would be helpful for you to say explicitly
what you mean by "an average", in particular what sorts of weights
w_i you allow.
>>
>> >
>> >As some of us on this side of the pond might say, "you're singing in
>> >the
>> >wrong choir and in the wrong church".
>> >
>> >The subject is the definition and meaning of the word "MODE",
>> >not pie a la mode or mode of transportation. It's the statistical
>> >NOUN which is a summary statistic for a set of numbers. or a
>> >description of thre relative maximum of a probability density function.
>> >
>> >-- Reef Fish Bob.
>
>This being the end of another month -- I hope this tedious thread
>about the SIMPLEST statistical definition and notion of a MODE
>(that of a point of the HIGHEST FREQUENCY or a point of relative
>maximum in a continuous density) will end in a STATISTICAL group,
>after it has been ongoing for 15 MONTHS.
>
>Those who are still muddled in this simplest of the statistical
>concepts
>should take up another field or discipline OTHER than statistics.
>
>-- Reef Fish Bob.
>
>-- Reef Fish Bob.
>
Reef Fish
st...@mimosa.csv.warwick.ac.uk wrote:
> In article <1154355451.4...@m79g2000cwm.googlegroups.com>,
> "Reef Fish" <Large_Nass...@yahoo.com> writes:
> >
> >st...@mimosa.csv.warwick.ac.uk ha escrito:
> >> >>
> >> >>
> >> >> to be considered an "average"?
> >
> >You were ignored because it is IRRELEVANT to the meaning of the MODE.
**** nonsensical noise snipped ****
> >Whatever you consider to be an average, the statistical MODE is NOT
> >that defined to be any average.
> >
> >> When you say refer to an "average", do you allow anything other than
> >> the arithmetic mean, and if so, what?
> >> -- Many thanks, Ewart Shaw
> >
> >You should have reviewed what had gone long before you jumped in.
> >It allowed the geometric mean, harmonic mean, trimmed mean,
> >winsorized mean, etc., etc. Any mean or average you care to use,
> >the MODE is NOT one of them.
>
> You allow the Winsorized mean, but in another post you emphatically
> said ("NO") that the median is not an average.
Oh NO!!! Just when we are about to kill the headless snake that has
been wiggling around the MODE ... another headless snake reared it's
ugly head -- misunderstanding what a MEDIAN is!
> But for sample size
> 2k+1 or 2k+2, the k-times Winsorized mean is just the median.
You had better review whatever comic book you've been learning
your statistics as to what a "Winsorized mean" is!
Suppose you get these values AFTER trimming the outliers (however
many)
0 1 2 10 20 100 200
The MEDIAN is 10. The trimmed mean is 47.571.
> J.E.H.Shaw [Ewart Shaw] s...@uk.ac.warwick
> Department of Statistics, University of Warwick,
You are in a Department of STATISTICS in the UK? What's this
world coming to??? First the pitt psychiatry Department Ulrich
USED to claim to belong as a statistician. Then the University
of Helsinki showed up with Anon Bob O'Hara. And now Ewart
Shaw of the University of Warwick in the UK!
This is a WORLD EPICDEMIC in statistical IGNORANCE!!!!
-- Reef Fish Bob.
st...@mimosa.csv.warwick.ac.uk
"Reef Fish" <Large_Nass...@yahoo.com> writes:
>
>st...@mimosa.csv.warwick.ac.uk wrote:
>...
>> >You should have reviewed what had gone long before you jumped in.
>> >It allowed the geometric mean, harmonic mean, trimmed mean,
>> >winsorized mean, etc., etc. Any mean or average you care to use,
>> >the MODE is NOT one of them.
>>
>> You allow the Winsorized mean, but in another post you emphatically
>> said ("NO") that the median is not an average.
>
>Oh NO!!! Just when we are about to kill the headless snake that has
>been wiggling around the MODE ... another headless snake reared it's
>ugly head -- misunderstanding what a MEDIAN is!
>
>
>> But for sample size
>> 2k+1 or 2k+2, the k-times Winsorized mean is just the median.
>
>You had better review whatever comic book you've been learning
>your statistics as to what a "Winsorized mean" is!
>
>Suppose you get these values AFTER trimming the outliers (however
>many)
As you probably know, Winsorizing is not the same as trimming.
>
>0 1 2 10 20 100 200
You give 7 values. Your original data-set obviously had
sample size 2k+1, for some k. After Winsorizing k times,
you get the adjusted "sample" 10 (2k+1 times).
Thus, as I said, the k-times Winsorized mean is the median.
>The MEDIAN is 10. The trimmed mean is 47.571.
>
>> J.E.H.Shaw [Ewart Shaw] s...@uk.ac.warwick
>> Department of Statistics, University of Warwick,
I was not looking for a definition of "average" that includes
(say) the mode, or the median, or a trimmed or Winsorized mean.
All I asked was for you (RF) to say exactly what you meant by "average".
Then it would have been clear why the mode and the median aren't
averages, but the trimmed and Winsorized means are.
> [rest snipped]
Reef Fish
s...@mimosa.csv.warwick.ac.uk () wrote:
> In article <1154450242.2...@75g2000cwc.googlegroups.com>,
> "Reef Fish" <Large_Nass...@yahoo.com> writes:
> >
> >st...@mimosa.csv.warwick.ac.uk wrote:
> >...
> >>
> >> said ("NO") that the median is not an average.
> >
> >Oh NO!!! Just when we are about to kill the headless snake that has
> >been wiggling around the MODE ... another headless snake reared it's
> >ugly head -- misunderstanding what a MEDIAN is!
> >
> >
> >> But for sample size
> >> 2k+1 or 2k+2, the k-times Winsorized mean is just the median.
> >
> >You had better review whatever comic book you've been learning
> >your statistics as to what a "Winsorized mean" is!
> >
> >Suppose you get these values AFTER trimming the outliers (however
> >many)
>
> As you probably know, Winsorizing is not the same as trimming.
Of course. But the PRINCIPLE remains the same. After trimming OR
Winsorizing the means are still NOT the MEDIAN as you falsely
claimed ... except for very special sets of data.
>
> >
> >0 1 2 10 20 100 200
>
> You give 7 values. Your original data-set obviously had
> sample size 2k+1, for some k. After Winsorizing k times,
> you get the adjusted "sample" 10 (2k+1 times).
> Thus, as I said, the k-times Winsorized mean is the median.
>
> >The MEDIAN is 10. The trimmed mean is 47.571.
> >
> >> J.E.H.Shaw [Ewart Shaw] s...@uk.ac.warwick
> >> Department of Statistics, University of Warwick,
>
> I was not looking for a definition of "average" that includes
> (say) the mode, or the median, or a trimmed or Winsorized mean.
> All I asked was for you (RF) to say exactly what you meant by "average".
> Then it would have been clear why the mode and the median aren't
> averages, but the trimmed and Winsorized means are.
I've said that in every possible way it could be sad.
You are just like a little gerbil now ... going round and round and
round in your little circular cage when you should have stood
still and would have seen the perfectly obvious anwer to your
own fallacy.
>
> > [rest snipped]
> --
> J.E.H.Shaw [Ewart Shaw]
> Department of Statistics, University of Warwick,
Unbelievable! Are you a student (about the fail in the program)
or are you actually dispensing some of your erroneous ideas to
some innocent souls in that Department?
-- Reef Fish Bob.