Description of MACD - request for feedback/comments
Jim Cochrane
I recently spent a good bit of time figuring out and documenting in detail
the MACD indicator and how it works, including several different ways of
using the indicator. I'm posting it here to elicit feedback and comments
from knowledgeable readers of this group - especially, any corrections of
errors in the document are welcome; and hopefully it will provide some
good insight into how to use MACD for others reading this newsgroup. I'm
not aware of any other description of MACD that goes into this kind of
detail.
Also, probably most people will find the concepts in this document hard to
visualize without some actual charts. (I know that I would find it hard.)
Because of that, I would recommend reading it while using a charting program
that provides the indicators described in the document. It also can be
helpful to create fake smoothed data to help see the relationships
between functions better, as mentioned in the document. If anyone would
like me to email them the data files that I created for this purpose, let
me know and I'll send them to you.
Thanks.
-----------------------------------------------------------------------
Moving Average Convergence-Divergence
The Moving Average Convergence-Divergence (MACD) indicator consists
of two functions: the difference between two moving averages (which I
call the MACD Difference) and a moving average of the MACD Difference
(which I call the MACD Signal Line). The MACD Difference is calculated
by starting with two exponential moving averages (of closing prices)
of different lengths. For example, for daily data, a 12-day exponential
moving average (EMA) and a 26-day EMA are often used. The MACD Difference
is the result of subtracting the larger EMA (26-day in the above example)
from the smaller EMA (12-day). [For a description of (EMA), see "Trading
for a Living", page 122.]
The MACD Signal Line is simply an EMA of the MACD Difference. A common
length for the MACD Signal line is 9 periods.
There is more than one way to use the MACD indicator for trading signals,
but a common way is to regard the crossing of the MACD Difference line
above the MACD Signal line as a buy signal and MACD Difference line
crossing below the MACD Signal line as a sell signal.
The Logic Behind MACD
The reason MACD works is that it is based on the different responses of
two moving averages of different lengths to a change in the trend of a
price graph. When the price trend changes, the two averages will cross.
This phenomenon provides the basis for a trading signal.
For example, when a price graph is in a steady uptrend, the shorter
moving average, since it is more responsive to changes in the price,
will be closer to and just below the price, whereas the longer moving
average will be further away from, and also below, the price.
If at a certain point the price graph reaches a peak and then goes into
a downtrend, the two moving averages will follow this movement, lagging
behind a little. Both averages will, soon after the price reaches its
peak, cross above the price and will remain above as long as the price
remains in a downtrend. However, since the shorter moving average is
more responsive to changes in the price graph, it will follow the trend
change more closely than the longer moving average and thus will end up
crossing the longer moving average as it turns down to follow the price.
During the downtrend the shorter moving average will again end up
being closer to the price than the longer moving average, although this
time they will be above the price graph. (It may be helpful for you
to draw out this example on a piece of paper to better visualize it.)
When the price graph changes from a downtrend to an uptrend, the same
event will occur, only in reverse - that is, both moving averages will
move from above the price graph (during the downtrend) to below it (when
the uptrend occurs). Again, the two averages will cross, just as they
did when the price changed from an uptrend to a downtrend.
Since this crossover effect happens whenever the price graph changes
from an uptrend to a downtrend and vice versa, this phenomenon can be
used as the basis of a trading signal that indicates a trend reversal.
(One could point out that the crossing of a moving average over the price
could also be used as a signal; but this signal is not as good as the
moving average crossover signal because the price can be volatile and
thus give false signals, crossing over the average and then back again.)
The crossover of the short and long moving averages is the same as the
crossover of the MACD Difference above and below 0 (since when the two
averages cross, the difference between them will be 0). Thus the MACD
Difference line crossing above 0 (caused by the short average crossing
from below to above the long average when the price changes from a
downtrend to an uptrend) can be regarded as a buy signal and the MACD
Difference line crossing below 0 (caused by the short average crossing
from above to below the long average when the price changes from an
uptrend to a downtrend) can be regarded as a sell signal.
There is a problem with using the MACD Difference line crossing 0 as a
signal: It tends to occur too late, compared to the timing of the price
trend reversal, to provide a useful signal. Let us further investigate
the nature of the MACD Difference line to see how it can be used more
effectively. To do this, we will compare it with the slope of the price.
(A good way to do this comparison is to create artificially smoothed
price data, so that the price is less volatile. This will make it
much easier to see the relationships described below. Also, using the
slope of a moving average instead of the slope of the price will further
decrease the volatility, making it easier to compare the two functions.)
If you choose a stock and compare the graph of the slope of its price
with its MACD Difference, you will notice that the behavior of these two
functions is similar (though with a slight lag for the MACD Difference,
since it is delayed by using moving averages): when the price has a
constant slope (moves in a straight line), both functions (the slope
of the price and the MACD Difference line) form a horizontal line;
when the slope of the price changes at a constant rate, both functions
form a straight line that slopes upward if the price graph is rising
and downward if the price graph is falling; when the slope of the price
graph changes at an increasing (i.e., non-constant) rate (for example,
increasing by 1, then by 2, then by 3 - rather than by 1 each period),
the slope of both lines changes at a constant rate.
[Why does the MACD Difference line have behavior that is similar to the
slope of the price graph? Essentially it is because when the price
is moving in a straight line, the two moving averages used to create
the MACD Difference will remain the same distance apart - thus the MACD
Difference will remain the same, making a horizontal line. When the price
is steadily increasing in slope when going up or decreasing in slope
when going down, the gap between the two moving averages will increase
steadily, resulting in a graph of a straight line with a positive slope
when the price is going up and a negative slope when the price is going
down. No doubt there is a much more precise mathematical explanation
of this phenomenon, using calculus, but although I may be capable of
figuring out this explanation, it will take much more time than I am
willing to invest, and the resulting explanation will probably only be
useful to those who have a good understanding of calculus.]
For those who are not mathematically inclined, the description above
may sound esoteric, but this behavior has a very useful application.
When the price reverses in trend from an uptrend to a downtrend, the
slope of the price and the MACD Difference line both begin to flatten
out and then decrease before the price reaches its peak. When a price
reverses in the other direction - from a downtrend to an uptrend,
this principal applies in reverse: the slope of the price and the MACD
Difference line will reach a bottom before the price reaches its bottom.
Although it is hard to see this effect on a typical chart, with volatile
data, it still does occur, and thus will provide a useful signal: When
the MACD Difference line reaches a peak (or, said another way, when the
slope of the MACD Difference line crosses below 0), this tends to give
a sell signal before the price has reached its peak; and when the MACD
Difference line reaches a trough (the slope of the MACD Difference line
crosses above 0), this tends to give a buy signal before the price has
reached its bottom. Since MACD Difference is derived using moving averages,
it tends to lag behind the slope of the price; but, since the slope
of MACD Difference crossing 0 is based on an event that occurs before
the price reaches a peak, it provides a signal that is not as late
as the MACD Difference line crossing 0 and is usually prompt enough to
provide a useful signal.
Note that using the slope of the price crossing 0 as a trading signal
could be used in the same way as the slope of MACD Difference. However,
the slope of the price will tend to be volatile and tend to produce too
many signals to be useful. (On the other hand, using the slope of a
moving average of the price may produce acceptable signals.) Using the
slope of MACD Difference crossing 0 as a trading signal provides a smoother
indicator that produces less false signals than the slope of the price.
But it still tends to produce more frequent signals than one would prefer,
especially when analyzing data for long-term trends.
Since the MACD Signal line is a moving average of the MACD Difference,
its movement basically follows the MACD Difference line, though with less
volatility. The standard signal of the MACD Difference line crossing
over the MACD Signal line provides a very similar signal to the slope
of MACD Difference crossing 0, but with a smaller number of signals.
The signal is similar to the slope of MACD Difference crossing 0 for
the following reason: The slope of MACD Difference crosses 0 when the
MACD Difference line reaches a peak or a trough. Since the MACD Signal
line is a smoothed version of the MACD Difference line, with a slight
lag, only when, or soon after, the MACD Signal line reaches a peak or
trough, will the MACD Difference line, which reached its peak or trough
before the MACD Signal, cross the MACD Signal line. (You can verify this
visually by examining the MACD indicator graph for several stock charts.)
Since the MACD Signal line is smoother than the MACD Difference line, it
will not have as many peaks or troughs and thus will give fewer signals
than the MACD Difference and a larger percentage of those signals will
be reliable than the MACD Difference signals. The disadvantage is that
there is a further lag and thus these signals are later than treating
MACD Difference reaching a peak or trough as a signal. In practice, I
find that this delay does not invalidate using MACD Signal for long-term
trend reversal signals. For short-term signals, the delay may be too
great for the signals to be useful. (Below, I will call this signal of
the MACD Difference line crossing over the MACD Signal line the "MACD
Crossover signal".)
Another use of the MACD Crossover Signal is to detect when the price
trend strengthens - for example, when the price changes from a slight
uptrend to a steep uptrend, or when the price resumes an uptrend after
going sideways for a while. The reason this works is that if the slope of
the price is at first constant, the MACD Difference line is horizontal;
when the slope of the price then begins increasing, the MACD Difference
line begins to slope upward. This movement of the MACD Difference line
will often create a trough when the MACD Difference line begins to
slope upward as the price increases in slope. This can be detected
as a signal: the slope of the MACD Difference line will cross above 0
and this will often also produce an MACD Crossover Signal. Of course,
this also provides a signal in the reverse case - when the price is in
a steady downtrend and begins falling at a faster rate.
As an example of using the MACD Crossover Signal to detect the
strengthening of a trend we can look at the weekly chart of SUNW in 1999.
Two buy signals occurred in the summer - one in early July and one in
late August. In April, the uptrend that had begun in October of 1998
relaxed a bit, going sideways for a few months. The first MACD Crossover
buy signal functioned as an alert that the uptrend had resumed in June.
The second signal functioned as an alert that the uptrend had strengthened
in August. Another example is the weekly chart for ENE in December of
1999. The MACD Crossover buy signal in late December gives an alert of
the resumption of the long-term uptrend after a few months of sideways
movement.
Other Indicators Based on the MACD Signal Line
Other effective indicators can be derived from the MACD Signal Line.
One useful indicator is the Slope of the MACD Signal Line and an
effective way to use this indicator is to regard its crossing above
0 as a buy signal and below 0 as a sell signal. This gives a signal
that is very similar to the MACD Crossover Signal - the MACD Crossover
Signal occurs shortly after the MACD Signal Line reaches a peak or a
trough. Since the Slope of the MACD Signal Line crosses 0 when the
MACD Signal Line forms a peak or a trough, this signal will occur a
little before the MACD Crossover Signal and thus will increase the
chances of catching a trend near the beginning. Like the MACD Crossover
Signal, I find the Slope of the MACD Signal Line crossing 0 to be a
good way of identifying long-term trends with weekly data.
Although I have not tried it yet in practice, a signal based on the Slope
of the Slope of the MACD Signal Line crossing 0 appears to be a good
signal to use to identify the end of a short-term trend that is going
against the prevailing long-term trend. For example, if a stock is in
a long-term uptrend and, within that uptrend, is in a short-term downtrend
or a sideways trend, the Slope of the Slope of the MACD Signal Line
crossing 0 often provides notice that the short-term trend is about to
reverse and join the prevailing long-term trend. Of course, this provides
a good signal for entering a trade.
In addition, the Slope of the Slope of the MACD Signal Line crossing 0 can
provide a good signal for exiting a trade. For example, if you are long
a particular stock and the current short-term uptrend for the stock is
about to end, which for some trading systems is a good exit point, the
Slope of the Slope of the MACD Signal Line will often cross below 0 just
as, or just before, the stock reverses into a short-term downtrend.
Calculus and MACD Signals
All of the signals based on MACD described above can be described using
the principal of derivatives from calculus. In calculus, what is called
the first derivative of a function (line in TA terms) is simply the slope
of the function. If you graph this slope, as we discovered above, the
graph will cross the 0 line whenever the underlying function reaches
a peak or a trough. When the underlying function is price, this provides
a signal that the price is reversing its trend. As we saw above, the
MACD Difference line behaves similarly to the slope of the price and thus
the MACD Difference line crossing 0 indicates that the price has reversed
trends.
The second derivative of a function in calculus is the slope of the slope
of a function. In the description above, the MACD Crossover signal and
the Slope of the MACD Signal Line crossing 0 both provide the behavior of the
second derivative of the price crossing 0. Since the MACD Difference
behaves like the slope of the price and the MACD Signal Line is a smoothed
version of the MACD Difference, the slope of the MACD Signal Line essentially
provides the behavior of the slope of the slope of the price. (And since
the MACD Crossover signal happens close to the time of the slope of the
MACD Signal Line crossing 0, it effectively provides the same
second-derivative-of-price behavior.)
The second derivative of a function measures the acceleration of
the function. Note that here we are conceptually regarding the slope
of a function as its speed and changes in the slope as acceleration.
For example, when a function slopes upward in a straight line, it has a
constant slope, or constant speed. The graph of the second derivative
for a line with an upward constant slope is a horizontal line whose
vertical coordinate is 0, since the acceleration of such a line is 0.
(The graph of the first derivative - the slope - in this case will be a
horizontal line, since the slope is constant.) If the function changes
in slope - for example, becomes steeper - the function accelerates.
Since the second derivative measures acceleration, when a function
accelerates at a constant rate - increases in slope the same amount each
period -, the y-coordinate (vertical coordinate) of the second derivative
will be be constant, forming a horizontal line.
For the purpose of technical analysis, the important thing to ask is:
"What happens when the graph of the second derivative crosses 0?"
Since the second derivative (slope of slope of a function) measures
acceleration of the underlying function, when its graph is at 0 (the
vertical coordinate is 0), the underlying function is not accelerating -
it is going in a straight line, up or down or sideways. The reason
this event is important is that when a price graph follows a typical
trend cycle, at the beginning of a trend the graph will increase in
slope - accelerate. Then at some point before it reverses trends, its
acceleration will begin decreasing until it finally stops accelerating -
the slope will stop increasing; after that point the slope will start
decreasing until it reaches a peak or trough and starts going in the
other direction. The point at which the price graph stops accelerating
is the point at which the second derivative crosses 0. Thus this event
signals a trend reversal before it actually occurs.
Since the event of the MACD Difference line crossing 0 is similar to
the slope of the price graph crossing 0, it is equivalent to the first
derivative crossing 0, which happens when the price reaches a peak or
trough before reversing trends. Since the MACD Difference lags behind
the slope of the price, the MACD Difference crossing 0 tends to occur too
late to serve as a good trading signal. The slope of the MACD Signal Line
crossing 0 (as well as the slope of the MACD Difference line crossing 0)
is equivalent to the second derivative crossing 0. Although this signal
is delayed because it is derived from moving averages, since it is a
second derivative signal it corresponds to the point where the price stops
accelerating before a trend reversal and thus is not as late as the
MACD Difference line crossing 0. I have found it to work fairly well as
a signal for a long-term trend change.
Since the Slope of the MACD Signal Line is equivalent to the second
derivative, the Slope of the Slope of the MACD Signal Line is equivalent
to the third derivative. It is difficult to connect this function with
the behavior of the price chart; I believe that the third derivative will
be 0 when the rate of acceleration of the underlying function is constant.
However, I think it is easier to reason about the Slope of the Slope of
the MACD Signal Line by seeing it as the second derivative of the MACD
Signal Line. The Slope of the Slope of the MACD Signal Line crosses 0
when the MACD Signal Line stops accelerating in preparation for reaching
a peak or trough. Since the MACD Signal Line can be thought of as an
indicator of the momentum of the price (for example, it reaches a peak
as the price begins to loose momentum before reversing trends), the
Slope of the Slope of the MACD Signal Line crossing 0 can be thought of
as indicating when the momentum has stopped accelerating and is about
to reverse trends. This event provides a forewarning of a forewarning
of an event - that is, the Slope of the Slope of the MACD Signal Line
crossing 0 warns that the MACD Signal Line is about to change trends,
which warns that the price is about to change trends. Although, as
I explained earlier, I have not yet checked out the reliability of the
signal, I have examined the daily charts of several stocks and it appears
that the Slope of the Slope of the MACD Signal Line crossing 0 on a daily
chart tends to give a very good signal for short-term trends, often giving
at least one or two days advanced notice before the price trend reverses.
--
Jim Cochrane
j...@dimensional.com
winch_p
For someone who is new to the game as you claim, you sure are long winded.
BTW there are plenty of good books with easier to comprehend explanations
than you offer. Murphy, Schwager, Kaufman...
P
Jim Cochrane
analyzing MACD and its behavior to really try to figure out how and why
it works. Other authors give a pretty good explanation of MACD, but I
don't think they go into this kind of detail or depth. One can
probably use MACD without such in-depth understanding, but I think such
a detailed explanation can be helpful. For example, I didn't realize
until I wrote the description that the MACD difference line (and also
the MACD signal line) behaves the same way as the slope of the price,
thus tending to give signals ahead of the price. I found this insight
very helpful to my understanding of how and why the MACD indicator
works. (Perhaps no one else found it helpful, since your criticism is
the only response I received.)
Did you take the time to read it? You don't comment on anything except its
length, so I suspect that you didn't. I can't give much weight to
criticism from someone who hasn't read what he is criticizing.
[And as I said in my other reply, I'm not really new to TA, although, I
suppose you could say I'm relatively new - a little over 3 years.]
In article <38f5f2f4$0$27...@news01.syd.optusnet.com.au>,
--
Jim Cochrane
j...@dimensional.com
Don Cameron
that it helped you understand a tool that you might use in trading.
It is usually a better learning experience to work something out for
yourself rather than rely on someone else to spoonfeed it to you.
Whether anyone else got anything out of it doesn't really matter,
although it was good of you to share.
I cannot recall exactly what Murphy and Kaufnam said, but I do not
recall mention calculus in them, whereas I thinkyour post did. If
that helps put it in perspective for you you are ahead of the large
majority of amateur TA users, most of whom I believe use indicators
without really understanding them.
On 18 Apr 2000 16:25:07 -0600, j...@dimensional.com (Jim Cochrane)
wrote:
TBone
winch_p
I didn't coment because it was heavy going and it was late at night.
Nevertheless I kept a copy. Don's comment were very good. It seems to me
that traders are a strange group. Deep fundamental appreciation would be
great but the combination with other things such as when to pull the
trigger, money management etc means that having traded with MACD Histo and
made money they would feel why does anyone need to know about impulse (third
derivative). BTW last time I saw third deriv was in defence department
trying to get a rocket over a mast.
Kaufman and murphy give nice mums and dads explanation of use of MACD and so
long as you stick to the rules of use its adequate. For most knowing to take
the safety off, point and shoot is enough. Interior ballistics knowledge
could come in handy but very rarely. Better that when the rules breakdown
get the hell out of there.
BTW most MACD come with defaults. Always test your data and figure what the
defaults should be for the market you are in. They do change with time
also.
You obviously felt you gained from the experience. MACD certainly is a well
used and basic tool.
Good luck
P
Jim Cochrane
purposes was to help me to understand better how MACD works.
In article <2lrpfssdt93uuc6p5...@4ax.com>,
--
Jim Cochrane
j...@dimensional.com
Jim Cochrane
Thanks for the response - good comments and advice. I agree that being
a good trader requires good experience and discipline (money
management, not listening to one's emotions, etc.), as well as
understanding the tools - indicators, etc. - one is using. I certainly
wouldn't want to spend as much time and energy as I did to write that
analysis of MACD on all other indicators I consider using. (It was
enlightening to write it, but it was very time consuming and painful.)
Regarding the third derivative, I'm actually finding it quite useful in the
form of the Slope of the Slope of the MACD Signal line (that is, the 2nd
derivative of the EMA of the MACD Difference). It essentially turns out to
be the third derivative because the MACD Signal line behaves the same as
the first derivative of the price and therefore the second derivative of
the MACD Signal line behaves as the third derivative of the price.
I've found this to be a good indicator for the reversal of a short-term trend
going against the prevailing trend - a good point for entering a trade. It
also looks like a good indicator for exiting a trade when the short-term
trend reaches a peak (or trough for a short) before reversing the other
way. Credit for this discovery should go to Jack Hershey. It was from one
of his posts here that I caught on to the idea of using the Slope of the
Slope of MACD Signal line as a trading signal. So surpisingly, the third
derivative actually appears to be useful in trading.
In article <38fdc9eb$0$27...@news01.syd.optusnet.com.au>,
winch_p <win...@optusnet.com.au> wrote:
>Jim,
ln...@yahoo.com
Jack Hershey
I have three general comments to make and if they are redundant to prior
posts I apologize since I haven't been in MIT newsgroup for quite a while.
1. The MACD period setup is very important and it has to harmonize with the
pace of the market. gerald Appel orginated MACD with the periods cited
below. 12, 26 and 9. I recommend that you consider they were established
when securities were mostly related to a molecular busines world. Today
with the advent of bits in additiont molcules there is a possibility that
the periods my be changed. Over the years I have arrived at a selection of
5 , 13 and 6. If you try this modification out you will see it picks off
the 6 to 8 day natural half cycle of price rise on most quality stocks. The
12 26 and 9 misses usually.
2. I believe that a pair of MACD's is terrific for making money. How you
get to use a pair will surprise you a lot. I use two computers. My example
is the DJXX which I trade intraday. To do the normal intraday trends that
occur, you will usually trade about 5 to 7 times a day. i set up my trading
screen with 5 min bars and the MACD of 1. above. I look at four MACD
points; the fast line peak, the pair intersection and the individual lines
each crossing the "0". After the fourth signal it is pretty sure that a
trend is underway. To enhance getting into the trend earlier on a reliable
signal, I use a second MACD pair. On the other machine I set up 1min bars
and use the MACD settings of 1 above. These lines really give me an early
warning on the trend of the five min bars. In effect the "trading fractal"
of 5 minutes can be enhanced by looking at the MACD "harmonics" on the 1 min
bars. When you check this out you are going to be amazed. When the 5 and 1
move together the 5 min MACD is really shifting. When they move in
opposition you see how a lateral move in price comes about. So you can also
then see that the lateral will end and the trend resume shortly. It is one
of the most relaxing (tension.alleviating phenomena) I know of for the
intraday commodities trader.
3. Use the calculus related phenomena of first derivatives to determine the
MACD pace of the trend you are trading. Once you get to optimizing capital
appreciation you will only want to be in the highest rate of return portions
of trends, i. e., the enter late and leave early approach as you rotate from
entity to entity. This facet easily gives you another doubling of your ROI
per year.
I use a three element pinwheel of MLR's on price to back up the MACD mlr I
also construct which deliniates the equilibrium (slope) of the MACD*. The
MACD mlr period is 30 and the price pinwheel threesome is 40, 20 ,10
periods.
* You strive for high MACD values and a steep MACD MLR to max to rate of
capital appreciation.
<ln...@yahoo.com> wrote in message news:39079358...@yahoo.com...
Jim Cochrane
[Sorry for the late response. I had to grab your article from dejanews.com,
since our newserver expired it before I had a chance to reply.]
Thanks for your comments; they are very helpful and not at all redundant.
I have a couple questions and comments about your post, below.
Jack Hershey wrote:
> Hi lnsn and Jim
>
> I have three general comments to make and if they are
> redundant to prior posts I apologize since I haven't
> been in MIT newsgroup for quite a while.
>
> 1. The MACD period setup is very important and it has to
> harmonize with the pace of the market. gerald Appel
> orginated MACD with the periods cited below. 12, 26 and
> 9. I recommend that you consider they were established
> when securities were mostly related to a molecular
> busines world. Today with the advent of bits in
> additiont molcules there is a possibility that the
> periods my be changed. Over the years I have arrived at
> a selection of 5 , 13 and 6. If you try this
> modification out you will see it picks off the 6 to 8
> day natural half cycle of price rise on most quality
> stocks. The 12 26 and 9 misses usually.
I assume you are talking about daily (rather than weekly) data here.
Do you use this setting for the "slope of the slope of MACD signal line" or
for MACD crossover - or for all daily MACD settings?
I've discovered that stocks go down faster and steeper in downtrends than
in uptrends - and so far 5, 11, and 4 appears to work well for sell signals
for a downtrending stock with the event "slope of slope of MACD signal line
crosses below 0".
For buy signals on uptrending stocks, I've been using the standard 12, 26,
9 parameters (for "slope of slope of MACD signal line crosses above 0"), and
I think you are correct that they are not tight enough. I'll try your
suggestion of 5, 13, and 6.
Do you use the same settings for both buy and sell signals, or do you use
different settings?
Do you also use MACD on weekly data (e.g., for detection of long-term
trends) and, if so, what settings do you use for weekly MACD?
This is fascinating. Since I don't use intraday data (just end-of-day) at
this point, I will put this in my library of good techniques to
investigate, and try it out when I start using intraday data.
>
> 3. Use the calculus related phenomena of first
> derivatives to determine the MACD pace of the trend you
> are trading. Once you get to optimizing capital
> appreciation you will only want to be in the highest
> rate of return portions of trends, i. e., the enter late
> and leave early approach as you rotate from entity to
> entity. This facet easily gives you another doubling of
> your ROI per year.
> I use a three element pinwheel of MLR's on price to back
> up the MACD mlr I also construct which deliniates the
> equilibrium (slope) of the MACD*. The MACD mlr period
> is 30 and the price pinwheel threesome is 40, 20 ,10
> periods.
I'm having a little trouble following this - probably because of my
ignorance and inexperience. First - what is an MLR? What is the MACD mlr
you are referring to - is it the first derivative - slope of MACD? And are
you talking about the MACD difference line or the signal line (EMA of the
difference line)? I also don't understand what you mean by the 3-element
pinwheel and what the 40, 20, and 10 period settings apply to. Can you
explain this in more detail?
Apologies for my ignorance; and thanks for your patience and generosity (in
this and all your previous posts). Hopefully there are others following
this thread who will also benefit from your elaboration.
>
> * You strive for high MACD values and a steep MACD MLR
> to max to rate of capital appreciation.
<ln...@yahoo.com> wrote in message
news:39079358...@yahoo.com...
> I would like to get the paper. My Email id is ln...@yahoo.com. Thanks.
>
> Jim Cochrane wrote:
>
> > I recently spent a good bit of time figuring out and documenting in detail
> > the MACD indicator and how it works, including several different ways of
> > using the indicator. I'm posting it here to elicit feedback and comments
> > from knowledgeable readers of this group - especially, any corrections of
> > errors in the document are welcome; and hopefully it will provide some
> > good insight into how to use MACD for others reading this newsgroup. I'm
> > not aware of any other description of MACD that goes into this kind of
> > detail.
> >
> > Also, probably most people will find the concepts in this document hard to
> > visualize without some actual charts. (I know that I would find it hard.)
> > Because of that, I would recommend reading it while using a charting program
> > that provides the indicators described in the document. It also can be
> > helpful to create fake smoothed data to help see the relationships
> > between functions better, as mentioned in the document. If anyone would
> > like me to email them the data files that I created for this purpose, let
> > me know and I'll send them to you.
--
Jim Cochrane
j...@dimensional.com