Jack Hershey's Method Doesn't Work
Rob Mullin
speaking, buying at 0 and 7 and selling at 4 and 3 **DO NOT**
provide an edge. None of your scores are predictive. Period.
Also, why can't you provide simple, straight-forward
explanations? For example, instead of going into your elegant
Boolean binary decimal mumbo-jumbo, why couldn't you simply have
said: Overall score = (4*Price score) + (2* Volume score) + A/D
score?
The answer is because there's no meat behing your cryptic
writings and you need the smoke screen to hide the lack of
content.
Jack Hershey wrote:
>
> >
> > I believe in the past you stated here that you use A/D and
volume as a
> > scoring system. Are these the equations you are refering to?
>
> Yes i have created a scoring system using the three most
significant
> trading variables.
>
> with three variables and a Boolean approach (binary math) you get
scored
> from 0 to 7 digitally. I score binary and convert to digital. if
it is up
> or A score 1 if it is down or D score 0. order the scores:
> price:volume:A/D. you have a # 001 or 101, respectively they are
1 and 5
> decimally speaking. stocks count down
> 7,6,5,4,3,2,1,0,7,6,5,4,3,2,1,0,7,6,5etc
>
> It is elegant to see. Fifth graders quickly learn charting. Buy
at 0 and
> 7, hold through 6,5,4 and sell at 4 and 3. 0,7 is theswing
through the
> trough and 4,3 is the swing over the peak.
>
> So you then know three things: 1. Where the stock is in the cycle,
> 2. What is coming next,
> 3. And how fast the cycle moves.
>
> This tends to take the mystry out of trading quite a bit. Draw a
saw tooth
> of a stock going up and down and fill in the numbers. score yours
and see
> how random your investments are.. quite a revelation. Buy at 0,7
and sell
> at 4,3. hold across 7,6,5,4 don't own anything that is going
down
> 3,2,1,0. This is not complicated.
>
> Underneath it all is the P,V relationship. I love it because it
is an
> elegant expression of a pragmatic theory from the thirties.
because fifth
> graders can tell if a switch is on or off, and then convert three
binary
> numbers to a decimal, the two steps conveniently allow them to
paper trade
> stocks like little financial sports.
> they see the P,V relation in no time then they start inventing
things. It
> is a great relief to children to learn that making money in
investing is a
> logical process. And that there are many logical processes they
can
> improve upon. For stuck adults with all the myths that abound it
is more
> difficult. For high school kids that understand calculus it is
even more
> exciting.
>
> >
> >
Harla Yesner & James Fowler
First of all, let me thank Jack for his excellent summary in 10+ parts. I'd also like to
thank snarley for posting a numerical/statistical summary of what he perceived is the
approach Jack uses. These are all bits and pieces I have put together studying Jack's
writings over the past few weeks.
However, Rob is right. I feel a little bit like a biblical scholar at this point, trying to
pinpoint what certain terms mean and get everything into a tight mathematical system. In
fact, I asked Jack the exact same question Rob did about the scoring [does Overall score =
(4*Price score) + (2* Volume score) + A/D?].
I have to admit that after about two weeks it all started to make sense to me, or the logic
anyway. However, I still have not managed a successful simulation based on the model he
presented (I can get positive profits but not with statistical certainty). Granted, I am
working in futures, but there should be very little that is different for the futures market
(shorter cycles and margin multiples).
With this new information from Jack I have some ideas for what I am doing wrong in terms of
implementing a model similar to his, but I have reached the "I'm pulling my hair out of my
head" stage. I have faith. I believe. I just can't confirm.
Specific problems that I am having relate mostly to smoothing vs. lag. The data can be quite
lumpy at times, especially volume (I am looking at tick volume). I need first and second
derivatives to be as current as possible, but I also need a nice smooth curve to avoid false
signals. The longer the lag, the smoother the curve. But I am still working on it.
I'm also unsure if I am calculating A/D correctly. I add volume when there is an uptick,
subtract it on a downtick, and do nothing when the tick moves sideways (though I am using 1
minute data, so I don't get tick to tick which might be my problem).
Finally, another question relates to fractal levels. Do you find that the same proportions
hold true for the indicators for each fractal level. E.G. if I used an 16 day moving average
of volume for stocks with a cycle of 8 days, I might use a a 60 minute moving average for a
futures contract with a cycle of 30 minutes.
Once again, thanks to each of you for all the great posts,
James
------------------------------
James Fowler
Department of Government
Littauer Center
Harvard University
Cambridge, MA 02138
Phone: (617) 661-6850
e-mail: jhfo...@fas.harvard.edu
Jack Hershey
My appologies for not being responsive to you. I have just completed a
writing and through its preliminary circulation for comments, numerous
extensive quotes have been extracted and used to deal with a team working
on a contemporary problem. I am having a dickens of a time formalizing
giving my permission for what has already thoughtlessly and inappropriately
transpired. It bodes well for the efficacy of the work , however..lol....
This is a terrific post that will give us a lot of progress in adding
statistical support for the investment approach at hand. Your concerns and
focus are right on.
I will parse the posting below and we can continue the thread here to
benefit others and privately to benefit ourselves when it might get boring
to most others.
I am also going to extract from this several good points for organizational
purposes. I will put these extractions into subordinate pieces to the ten
parts already in place.
Harla Yesner & James Fowler <jhfo...@fas.harvard.edu> wrote in article
<37EF29AF...@fas.harvard.edu>...
> Hi Rob, Jack, and the rest of MIT
>
> First of all, let me thank Jack for his excellent summary in 10+ parts.
I'd also like to
> thank snarley for posting a numerical/statistical summary of what he
perceived is the
> approach Jack uses. These are all bits and pieces I have put together
studying Jack's
> writings over the past few weeks.
Great. I am going to get together with snarley soon. He lives in between
me and the glider I am currently using. And the skyport owner has his
machine set up with these overwrites as does my flight instructor (there is
always room for improvement). Snarley didn't sort out exactly what I said
in some instances because of my poor articulation. But he did get 90% of
it.
>
> However, Rob is right. I feel a little bit like a biblical scholar at
this point, trying to
> pinpoint what certain terms mean and get everything into a tight
mathematical system. In
> fact, I asked Jack the exact same question Rob did about the scoring
[does Overall score =
> (4*Price score) + (2* Volume score) + A/D?].
yes. I am hoping the Word version of this will be exportable in the near
future. It has color graphics supporting it and it works like a whiz for
those I mentor here.
>
> I have to admit that after about two weeks it all started to make sense
to me, or the logic
> anyway. However, I still have not managed a successful simulation based
on the model he
> presented (I can get positive profits but not with statistical
certainty). Granted, I am
> working in futures, but there should be very little that is different for
the futures market
> (shorter cycles and margin multiples).
This is the key thing on the table. All most everyone makes profits and
usually better profits than before. I appreciate and agree with the notion
that a statistical backup for a method really puts the icing on the cake.
Since this approach is now generalized, I feel as you do that working on it
in the area you are (futures intraday trading) is as good a place as any
the achieve a statistically significant measure. In stocks, the universe
departs from a stratified random sample of all stocks and that is a factor
designed to minimize risk. In futures, there are two factors that come
into play: the breadth of the market and the timeliness of each of the
relatively few formalized future entities, granting that all equities can
be "played" if enough players are available. I am using this approach on
active DJXX's which roll over every 90 days. I am monitoring all other
futures indexes and I can mentor those who choose whatever index they
prefer.
>
> With this new information from Jack I have some ideas for what I am doing
wrong in terms of
> implementing a model similar to his, but I have reached the "I'm pulling
my hair out of my
> head" stage. I have faith. I believe. I just can't confirm.
Faith, belief, and making profits go along way towards confirming because
they set up an ambiance whose subtle bias that leads to results instead of
not completing the work.
>
> Specific problems that I am having relate mostly to smoothing vs. lag.
The data can be quite
> lumpy at times, especially volume (I am looking at tick volume). I need
first and second
> derivatives to be as current as possible, but I also need a nice smooth
curve to avoid false
> signals. The longer the lag, the smoother the curve. But I am still
working on it.
There are several ways to deal with this. The underlying facet is to be
sure to have the data scenerio you are using be in sync. The reason I
chose to use the natural cycle (6-8 days) here was because everyone is
locked into EOD data and that is the optimum duration in that setting. I
don't like EOD data because it doesn't depict the market well. But we are
stuck with it just like how we are stuck with the inane simplistic level of
the media reporting out the financial market news. Believe me someday I
am going to go to one of those local TV things with just a camera and
lights and get a slot for fifteen minutes and just put those guys out of
business. I look like Sean Connery and Sound like David ________.
The first and second derivatives are right on for shifting lagging data to
leading data. and I know there is a predicting trap there...so be it.
If you can easily do it. Compare the three duration (paces) of intra day
futures index trading. fast (5 to 30 seconds - i call it tight); medium (5
to 30 min - I call it drift); and slow (30 min to 1 hour - I call it
trend).
Use as your data scenario: for fast: line, realtime every data item; for
medium: 5 (preferred) and 1 min bars and 1 min ancillary data; and slow: 30
(preferred and 10 min bars, and 5 min ancillary data.
I always am swinging during the day from one pace to the others by dragging
my feet from the slowest possible to faster paces. As this stuff goes
electronic, It will be easier.
You will get best statistical significance with the slow I think.
Next do this. Cut out the A/D and use the cruder four part cycle, so you
are pristine and clean vis a vis the P, V relation. I use A/D because it
is so powerful on the entry and exit and it is why contrarians are so
successful as gleeners. The A/D reversals mid trends are like breathing is
to the heart beat of trading.
>
> I'm also unsure if I am calculating A/D correctly. I add volume when
there is an uptick,
> subtract it on a downtick, and do nothing when the tick moves sideways
(though I am using 1
> minute data, so I don't get tick to tick which might be my problem).
The finesse required here gets almost to a level of being arbitrary. Good
A/D measures are complex and therefore take sufficient duration. For EOD
the first derivative of Worden's Bof P is great. Worden doesn't have the
math for the first derivative. Talking with their senior programmers is
pleasant as all get out but they are ruled by getting subscribers for their
supply of data. Another slow one is IBD's five segment A/D. You need to
get the tick version working for you. That is real important. My focus
when the timing gets brisk is to know from my market matrix that the market
always chooses its new operating point as one of the adjacent cells. I am
not going to get into that level of market thoery and practice here, but
understanding how the market moves from one operating point to another is
almost the most cardinal principal available to us today. If you choose to
score intraday one line charts, you will come to a spine tingling
conclusion. I chose for one day (after a day of observation - in other
words a cold start on a stock) to post intra day in raging bull. I did 30
posts. simply oriented as to what would happen 5 to 15 minutes in advance
of the market using "fixed" trailing indicators and all this stuff.
because of the uniquness of the day, several posters more than doubled
their position with as few as three trades only trading one way (long in
all cases). The day was 27 AUG 99 and the posting Board was SNMM. I did
only a few posts after that to answer specific questions. I posted on
Clearstation under SNMM during that period to answer Q's because i was
watching SNMM there for a friend who uses Clearstation.
Those 30 posts show how knowing that the operating point of the market is
changing can affect favorable the results thia approach gives. It is not
necessary to change thusly. If we can get the stats down on one level of
observation for the duration of cycles, then we can consider looking at how
in the harmonics of overlain frequencies the traders drive the market to
carrying the melody in different harmonics. I am sitting on a keg of
dynamite so to speak because I know all of this can be reduced soon to
automated computerized trading. The statistical significance of the
approach is a strong key to acceptance.
>
> Finally, another question relates to fractal levels. Do you find that
the same proportions
> hold true for the indicators for each fractal level. E.G. if I used an
16 day moving average
> of volume for stocks with a cycle of 8 days, I might use a a 60 minute
moving average for a
> futures contract with a cycle of 30 minutes.
Terrific Q.!!!! You will like the answer but not the caveat. Sure it is
exactly as you say just allow a little organic effect for transients that
enter. They will show up to tell your their roles. Like trace minerals in
organic farming. They allow the soil biota to prep the plant nutrients
(chelating phenomena). Scalpers are like soil biota. Caveat: if you help
yor self too much, i. e., choose values to smooth blah blah, you will not
be in sync. What Pring did for Wilder recently on the RSI, brought it to
where I use it so it works for the natural cycle. Notice how I adjusted the
MA of the short MACD. The key is to keep the signal indicator duration at
a pace (length) which is less than the 1/2 cycle you are examining. This
is the major failure of the default setting on almost ALL SOFTWARE PACKAGES
regardless of the marketing strategies devised to determine their price to
make money for the software manufacturers.
I use a 5 MA on volume for the natural cycle of 6-8 days. This is a value
as great as i could make it for smoothing and it is "inside" the half cycle
duration. For it to cross something, the sluggish 30 MA of volume is
there. But to "fix" it to cure the inherent lag i use derivatives of the
rough 5 MA curve.
We are facing such sweeping things on the market horizon that we may come
to the end of EOD data.
division by two, as in 1/8ths to 1/16ths to 1/32nds is giving away to
...the decimal system. There are thoughts about the length the market is
open each day... Mathematically speaking, it certainly will induce another
wave of analysis techniques.
If we can go just a little further here, the staistical backup for this
will be refined fairly well. Then with a confirmation of the fact that it
is profitable, it can then be used by more young people who need it to
begin their lives in a more financially rewarding way.
If a lot of people copy cat this stuff for their existing audiences it will
help too.
Mike Brett
Can't say I care for the name of this thread but this is where the questions
are showing up so I'll parse mine in. They are mostly dumb ones; I hope you
don't mind.
Jack Hershey wrote in message <01bf0909$01ae0280$3f3184ce@default>...
ES99 is the arena I am currently playing in because of its level of
participation and the breadth measurements that the stock market provides.
Where and how do I sign up for mentoring? :)
>>
>> With this new information from Jack I have some ideas for what I am doing
>wrong in terms of
>> implementing a model similar to his, but I have reached the "I'm pulling
>my hair out of my
>> head" stage. I have faith. I believe. I just can't confirm.
>
>Faith, belief, and making profits go along way towards confirming because
>they set up an ambiance whose subtle bias that leads to results instead of
>not completing the work.
>>
>> Specific problems that I am having relate mostly to smoothing vs. lag.
>The data can be quite
>> lumpy at times, especially volume (I am looking at tick volume). I need
>first and second
>> derivatives to be as current as possible, but I also need a nice smooth
>curve to avoid false
>> signals. The longer the lag, the smoother the curve. But I am still
>working on it.
For intraday volume I currently rely on the eyeball technique until I can
fix my software. I can certainly see volume picking up during trends but I
cannot see the mode change signal and as James says the picture is very
lumpy. Although the faith is there the confidence factor is not high at this
point and I fall back on the interplay between up and down volume as a
confirming tool. I am using NYSE volume for this but I wonder if the futures
volume is where I should be.
>
>There are several ways to deal with this. The underlying facet is to be
>sure to have the data scenerio you are using be in sync. The reason I
>chose to use the natural cycle (6-8 days) here was because everyone is
>locked into EOD data and that is the optimum duration in that setting. I
>don't like EOD data because it doesn't depict the market well. But we are
>stuck with it just like how we are stuck with the inane simplistic level of
>the media reporting out the financial market news. Believe me someday I
>am going to go to one of those local TV things with just a camera and
>lights and get a slot for fifteen minutes and just put those guys out of
>business. I look like Sean Connery and Sound like David ________.
>
>The first and second derivatives are right on for shifting lagging data to
>leading data. and I know there is a predicting trap there...so be it.
OK. It's important for me to get the definitions down or I'm gone before I
start. When we speak of first and second derivatives we are referring to the
'fixed' indicator values? The MACD is the first derivative and the signal
line is the second? You are keying on the relative max and min of the second
derivative? I told you these questions would be dumb but once I get these
down I'll move on to Hadamard matrix analysis. LOL.
>
>If you can easily do it. Compare the three duration (paces) of intra day
>futures index trading. fast (5 to 30 seconds - i call it tight); medium (5
>to 30 min - I call it drift); and slow (30 min to 1 hour - I call it
>trend).
Are you monitoring all three paces at once or are you choosing one and what
are we looking for in our comparison?
>
>Use as your data scenario: for fast: line, realtime every data item; for
>medium: 5 (preferred) and 1 min bars and 1 min ancillary data; and slow: 30
>(preferred and 10 min bars, and 5 min ancillary data.
>
Jack, this last statement has me so confused I can't even formulate a
question. OK, I'll try. When analyzing the medium pace you chart 1 and 5
minute bars with all studies on price and volume? What do you mean by
ancillary data(macd's, rsi, ma's of volume)? The rsi 5,10 refers to a five
period rsi with 10 period smoothing ma?
>I always am swinging during the day from one pace to the others by dragging
>my feet from the slowest possible to faster paces. As this stuff goes
>electronic, It will be easier.
>
>You will get best statistical significance with the slow I think.
>
>Next do this. Cut out the A/D and use the cruder four part cycle, so you
>are pristine and clean vis a vis the P, V relation. I use A/D because it
>is so powerful on the entry and exit and it is why contrarians are so
>successful as gleeners. The A/D reversals mid trends are like breathing is
>to the heart beat of trading.
I am unsure of what you mean by the four part cycle here. An explanation
would be great.
I read this thread after you mentioned it in this ng. It was lots of fun as
I refered to a historical chart of the day in question. So you are
monitoring intraday volume using 5 and 30 ma's at the tick level? Am I even
warm here?
So the indicators have been fixed to work in the 6 to 8 day natural cycle.
How does this relate to intraday time periods? Are cycles of this length
predominant at the 3 paces mentioned above?
OK. here's that question again. Derivatives of the rough 5ma curve? Not sure
what you mean.
>We are facing such sweeping things on the market horizon that we may come
>to the end of EOD data.
>division by two, as in 1/8ths to 1/16ths to 1/32nds is giving away to
>...the decimal system. There are thoughts about the length the market is
>open each day... Mathematically speaking, it certainly will induce another
>wave of analysis techniques.
>
>If we can go just a little further here, the staistical backup for this
>will be refined fairly well. Then with a confirmation of the fact that it
>is profitable, it can then be used by more young people who need it to
>begin their lives in a more financially rewarding way.
>
>If a lot of people copy cat this stuff for their existing audiences it will
>help too.
>
>
>
That's it for the dumb questions. Can't say you weren't warned. Thanks again
Jack.
Deppest Regards,
Mike
Harla Yesner & James Fowler
Thanks for the thoughtful response. Mike has some very good questions (not
stupid at all) and I hope you will take some time to reply to his post.
For now I am puzzled about one thing you said:
> The finesse required here gets almost to a level of being arbitrary. Good
> A/D measures are complex and therefore take sufficient duration. For EOD
> the first derivative of Worden's Bof P is great. Worden doesn't have the
> math for the first derivative. Talking with their senior programmers is
> pleasant as all get out but they are ruled by getting subscribers for their
> supply of data. Another slow one is IBD's five segment A/D. You need to
> get the tick version working for you. That is real important.
This is not helping me program my simulation to calculate A/D! I'm not
familiar with Worden or IBD. Can you fill in details of how <mathematically> I
would calculate A/D in an appropriate way? Or at the very least provide a
source I can turn to?
James
Jack Hershey
A/D. The type of market, buyer or seller, is really important. A lot of
the Broker feeds include market buy/sell condition indicators. Where we
are here is getting into the way market indicators are developed.
Worden's B of P and MS are good and I will try to see if I can grab some
.TXT's that verbalize it for you. O'Neil's 40K broker service is feeding a
lot of funds now with what we want. I have a quid pro quo with one of his
protege's that I can tap.
Jack Hershey
Mike Brett <mbr...@panix.com> wrote in article
<7sp3qh$kqg$1...@news.panix.com>...
> Hi Jack,
>
> Can't say I care for the name of this thread but this is where the
questions
> are showing up so I'll parse mine in. They are mostly dumb ones; I hope
you
> don't mind.
I'll parse back to you.
> ES99 is the arena I am currently playing in because of its level of
> participation and the breadth measurements that the stock market
provides.
> Where and how do I sign up for mentoring? :)
> >>
we can do stuff here and by email at least.
>
> For intraday volume I currently rely on the eyeball technique until I can
> fix my software. I can certainly see volume picking up during trends but
I
> cannot see the mode change signal and as James says the picture is very
> lumpy. Although the faith is there the confidence factor is not high at
this
> point and I fall back on the interplay between up and down volume as a
> confirming tool. I am using NYSE volume for this but I wonder if the
futures
> volume is where I should be.
I would use the NYSE as a guide occassionally but focus on the index
primarily. Use the P,V relation as the immediate monitoring tool. Lumpy
is bad for him; for you it is a measure that you are getting signal well
above the noise level. Trading intrady requires you to anticpates and move
you trailing stops promptly when the pace (volume per unit time) picks up.
I track volume with a bar display. My line charts are paired (side by
side) for the NYSE --INDU and the DJXX for example. The line charts best
display formations and market pace. The bars I use change with the pace.
>
> OK. It's important for me to get the definitions down or I'm gone before
I
> start. When we speak of first and second derivatives we are referring to
the
> 'fixed' indicator values? The MACD is the first derivative and the signal
> line is the second?
No, I am using calculus short hand here. Think of the curve we are using
as the thing that will signal us.. The time rate of change of the curve is
called its slope, how it is tipping at any given time. estimate that with
a rule by placing it on the chart. This, in calculus is the first
derivative. If you plot the values of the slope and then repeat the same
thing for slope, i. e., taking the slope of that new graph you get the
sceond derivative. I do it mentally or invent a way to do it on the
machine. This is the same as the relationship of displacement, velocity
and acceleration.
We are using the MA of the short MACD as our indicator curve. most people
watch MA cross MACD as a signal. We watch the slope of MA go flat at the
top and act because after it goes flat it tips down to cross the MACD. The
result is acting earlier or looking at it another way the lag of waiting
for the crossing is taken out. By using the second derivative you wait for
it to fall to zero which is the same as the flat....... So if you watch it
go to zero, there is a time when it turns down to start to go to zero.
?That is its falt point. This is the exact time that the MA has reached
its maximum growth rate and is beginning to slow down. When the CD (
convergence/divergence ) MA starts to decay the price is beginning the end
of its trend. Now we have a leading indicator on the end of the trend
coming. thanks for reading through this.
You are keying on the relative max and min of the second
> derivative?
Yep, see above with the thought of what i said above..
I told you these questions would be dumb but once I get these
> down I'll move on to Hadamard matrix analysis. LOL.
Here is where you are. You are on the brink of something. I can't tell
you how little the distance is that you have to go. I have seen people go
thruogh this for both stocks and futures index trading. i have hours and
hours of people being taped in Q and A's and on the machine. Producers
have seen the tapes and say there is no way that the shear impact of the
recognition of understanding could be better displayed.
You are going to see it all fall in place and it is like driving a car.
Except the pace is like a bobsled race or glider acrobatics where you just
know you'll never loose it no matter how loud it gets at 5G's.
> >
> >If you can easily do it. Compare the three duration (paces) of intra
day
> >futures index trading. fast (5 to 30 seconds - i call it tight); medium
(5
> >to 30 min - I call it drift); and slow (30 min to 1 hour - I call it
> >trend).
>
> Are you monitoring all three paces at once or are you choosing one and
what
> are we looking for in our comparison?
Doing it all. They are harmonics on the waveform. I have a matrix of
cells. We will have several stops listed ahead and we will watch the
market move in a symphony from cell to cell telegraphing all the way.
It like going up vertically in a glider. There's no question it has
stopped moving and it won't fly backwards. LOL. so kick the rudder and
watch it just swing over sideways slicing the aire with the wing. finally
you are going straight down with not control whatsoever until you pick up
speed. you are just sitting there watch you double down reversal
accelerate like heck in the the profits because you knew the stop had to be
doubled by the action of the market (ship you're flying).
> >
> >Use as your data scenario: for fast: line, realtime every data item; for
> >medium: 5 (preferred) and 1 min bars and 1 min ancillary data; and slow:
30
> >(preferred and 10 min bars, and 5 min ancillary data.
> >
> Jack, this last statement has me so confused I can't even formulate a
> question. OK, I'll try. When analyzing the medium pace you chart 1 and 5
> minute bars with all studies on price and volume? What do you mean by
> ancillary data(macd's, rsi, ma's of volume)? The rsi 5,10 refers to a
five
> period rsi with 10 period smoothing ma?
>
Bar thuff is the price part of the chart.. ancillary is volume and the A/D
tick stuff.equivalents.
I have a level I monitor to check Bid/ask and sizes.
> >I always am swinging during the day from one pace to the others by
dragging
> >my feet from the slowest possible to faster paces. As this stuff goes
> >electronic, It will be easier.
> >
> >You will get best statistical significance with the slow I think.
> >
> >Next do this. Cut out the A/D and use the cruder four part cycle, so
you
> >are pristine and clean vis a vis the P, V relation. I use A/D because
it
> >is so powerful on the entry and exit and it is why contrarians are so
> >successful as gleeners. The A/D reversals mid trends are like breathing
is
> >to the heart beat of trading.
>
> I am unsure of what you mean by the four part cycle here. An explanation
> would be great.
A cycle with three variables is divided into 8 parts; with two variables
forur parts define it.
>
> I read this thread after you mentioned it in this ng. It was lots of fun
as
> I refered to a historical chart of the day in question. So you are
> monitoring intraday volume using 5 and 30 ma's at the tick level? Am I
even
> warm here?
Yes you get it. What happens is this. You scale it proportionally if you
wish. 5 bars and 30 bars at any pace you choose. Strange but it is the
same as looking at EOD bars even though it is intraday. The key is the
market pace. They are all there but at any given time one stands out.
>
> So the indicators have been fixed to work in the 6 to 8 day natural
cycle.
Yes that is the point of departure. It is the easy way to make money for
the beginning.
> How does this relate to intraday time periods?
they are microcosms of the 6 -8 day. And you just switch markets to have
one that gives you the 10% per cycle all the futures indexes do quite
nicely. The reason is not hard to figure out. It is a people thing.
Whereevr people are trading the way they trade is about the same.
Are cycles of this length
> predominant at the 3 paces mentioned above?
Yes James is working in the DJXX on this and the durations are as
mentioned.
> OK. here's that question again. Derivatives of the rough 5ma curve? Not
sure
> what you mean.
On price for any of the three paces. yes.
>
>
Mike Brett
Jack Hershey wrote in message <01bf0add$8d0309e0$9bad30d0@default>...
>
> snipped for brevity
Jack,
Thanks for your time and patience. I am in the process of digesting your
replies and I will continue to participate with more ideas and questions.
Once again I want to encourage you to continue with the postings and Q&A's
because what you are doing here is extremly helpful to myself and others.
Deepest Regards,
Mike
no...@nowhere.net
I echo Mike's sentiments. Thank you very much, Jack. Now would you be
kind enough to post a few graphs of price, volume and MACD
relationships? :-))
Jack Hershey
weekend. that way i can just keep rotating through it. You have the two
pairs of MACD's that will go with it.
MattZ
I loaded them in and they do appear to be 100% functional.
I only obtained 3 posts. I dont know if there were any others.
in the easy scan I over wrote the ones worden bros gave. the following
scans now appear.
aa stage 1 or 3
aa stage 2 or 4 ( high to low grid)
gainers over yesterday (5-10%)
gainers over yesterday >10%
gappers
low volume (mavv30-mavv5)
low volume ( 3.6 - mav 3)
low volume (6.3 - mav 3)
volume increase
volume old
I hope this is all.
mattz
--
------ICQ#42601178
The thing that hath been,it is that which shall be;
and that which is done is that which shall be done:
and there is no new thing under the sun.
Ecclesiastes 1:9 book of Genesis
..
Jack Hershey
and even is turns. It does all the stocks in the universe from high grid
to low grid
MattZ <ma...@pond.com> wrote in article <37F535F5...@pond.com>...
MattZ
Well I finally made it through all the postings Jack Hershey made in his
I-X series. I took my grand time reading those several times.
NOW; I have been collecting stocks with a high A/d and EPS (90+ on both)
float of 5-30mil price of 15-50$ (I think those are the number of the
top of my head) I have been rotating them through 2 watch lists in
tc2000 v4 and 3 (ive been using version 4 longer than 3 so It easier for
me right now to use vers 4's watch list. don't ask, I understand what im
doing) I have also coupled a 2 macd's with a RSI. I am looking for the
RSI to drop down through the macds (min and max)
I have discovered after several weeks that these stocks go into DU. I am
doing my best to understand DU. I have been doing this for a while now
and many of the stocks are rotating in and out of the scans Jack
provided. I do believe I understand if my 'interpretations' are correct.
Jack, If you have a few minutes, could you post a few examples of stocks
that you are tracking in various stages of the scoring? A real trade
example? I do believe I understand and a few examples would confirm what
I am looking at and interpreting correctly. A few examples (good and
bad!) will go a long way in my little pea brain.
Have you started a clearstation recommend list? That would be extremely
helpful. Not to coattail you but to further understand. I am almost
fully operational, I just getting the edges sharpened up.
ALSO what I am surprised about is that more people have not really
jumped on and started more with the thread Jack started.. The part on
risk reduction alone is worth a reread.
There are a few other small items ill toss in the mix, but that is
another time and post. ;)
I do thank you so much for your time !
Mattz
------ICQ#42601178
The thing that hath been,it is that which shall be;
and that which is done is that which shall be done:
and there is no new thing under the sun.
Ecclesiastes 1:9 book of Genesis
..
"Would you tell me please which way I have to go from here?'
"That depends a good deal on where you want to go,' said the cat."
Lewis Carroll, Alice's Adventures in Wonderland
winch p
For the sake of completeness, copies of Hersheys parts have been requested
from and sent by MattZ
P
MattZ <ma...@pond.com> wrote in message news:38167963...@pond.com...
MattZ
you can obtain them from deja news
can you post a small binary here?
mattz
--
------ICQ#42601178
The thing that hath been,it is that which shall be;
and that which is done is that which shall be done:
and there is no new thing under the sun.
Ecclesiastes 1:9 book of Genesis
..
He who has the truth and kept it,
Keeps what not to him belongs;
Keep a pearl from him who needs it,
And a fellow mortal wrongs.
W.D.Gann