MDA - Market Direction Analysis

[augieboo]

Tom, I have a little time this afternoon, so I'm going through my pile of "posts I've been meaning to respond to" and found yours re breadth indicators and your work on models. (I apologize for taking so long to answer, but please don't feel bad. It is usually the posts I find most interesting which are the ones I don't get around to for a long time, because I put them aside so I can really give them some thought, rather than just spewing off the top of my head. Then they scroll off the bottom of my monitor and aren't heard from again for a long, long, time.)

Anyhoo, you asked for my thoughts on your six behaviors (etc) exhibited by the market -- in particular you asked for my take on market internals. I'll give you those plus some other assorted bits of wisdom I've snatched from various sources lately.


==============================================================
1. Price Trends: Prices tend to trend. The best bet is
to stay on the right side of the trend.
==============================================================

Defining "the" trend is an impossible task, because the market may trend up, down, and sideways -- all at once -- depending on which time frame(s) you are looking at.

A tidbit from Alexander Elder:

[1] Pick your favorite time frame and call it "intermediate."

[2] Look at the next larger time frame (your nominal "long" time frame) and see which direction it is trending. (He uses only MACD and a single EMA for this.)

[3] Trade oscillator swings in your intermediate trend in the direction of your long term trend.

[4] Use your "short" time frame to help nail down entry, exit, and stop loss points.

Our thread founder and guiding light, LG, takes the concept of multiple time frames even further with a system I have dubbed "MTFA" for "multiple time frame analysis." (Please note that my ONLY contribution to MTFA was to give it a somewhat catchy name.)

You can find some posts explaining the basics of MTFA on my "junkpile" thread, which is #board-1220. Note also that Jim, aka 'hk2' of this thread is quite well versed in this technique.


========================================================================
2. Breadth Trends: Trends in breadth tend to mirror trends
in price. Beware of reversals when breadth trends reach
extremes, HOWEVER, because you can never know how high is too
high or how low is too low, don't buck the current trend.
Also - When the breadth trend diverges from the price trend,
there is a strong tendency for a reversal.
========================================================================

I would break this down a bit further.

[1] There are three types of "internal" information which I find quite valuable. In order of importance (in my system) they are:

a. new highs & lows (NHNL)
b. advancing & declining volume (AVDV)
c. advancing & declining issues (AIDI)

The reason that new highs & lows are so important is because those numbers tend to trend, even in their raw form, while raw AD numbers (whether volume or issues) are pretty indecipherable.

Once one has decided to use any internals data, the first task is to put it into some format in which a single data series can take into account both the ups (new highs, adv volume, adv issues), as well as the downs, (new lows, etc), while simultaneously taking into account the changes in the market over time. It is important to take all of these factors into account because:

-- the number of stocks in the Naz change a great deal over time, so comparing raw numbers of advancing or declining issues or new highs/lows over a long period of time gives false information. See #msg-760443 for a listing of changes in numbers of issues on the major indices over time.

- the percentage of issues which trade every day has increased greatly over the past couple decades, which will further skew raw numbers on A/D or NH/NL.

-- total market volume expanded massively during the 1990s and we have yet to figure out for sure if it is contracting on a long term basis or not. This would further skew data based on raw numbers.

Here are how Net New Highs, (NNH, i.e., new highs - new lows) look when calculated in each of three different ways and then charted:


1. Raw NNH:




2. NNH as a percentage of total issues traded:




NNH as a percentage of IA+ID




For my taste, the third method is best able to take into account all of the three problems I described, supra, so it is the method I use.


Okay, so, what about Issues and Volume? What should we do with these things?

Well, after trying out scores (if not hundreds) of combinations, my considered opinion is that both AVDV & AIDI are best handled through a summation index setup, after first being "netted" via the following formulas:

-- Volume

N%VA (net percentage volume advancing) = (VA - VD) / (VA + VD)

-- Issues

N%IA (net percentage issues advancing) = (IA - ID) / (IA + ID)


Once I get all of the above figured out, I combine all three types of data into a single indicator with three separate time frames. I give each time frame a moving average to serve as a trigger, fiddle with the numbers a little bit so that the data series do not plot on top of one another, and then chart them out to use as my main trend indicator.

Here's what the finished product looks like:




==============================================================
3. Price Momentum: Price momentum tends to lead market
prices. The longer the divergence between price and momentum,
the more likely a reversal will occur.
==============================================================

A key point to keep in mind is that bounded oscillators are best used in range-bound markets but become useless during strong trends, while unbounded oscillators are not nearly as useful during range bound markets but shine during strong trends. (Thanks to several posters who beat me over the head with this point enough times so that it finally sunk in.)


==============================================================
4. Breadth Momentun: Breadth thrusts (short-term periods of overwhelmingly positive breadth) tend to have bullish implications for intermediate/long term prices. Selling climaxes(short-term periods of overwhelmingly negative breadth) also tend to be bullish EXCEPT the market may experience more down side action over the short term.
==============================================================

My long-term analysis of NAZ internals, plus the charts of the NYSE summation index going back to the 1920s that I purchased from DecisionPoint.com all validate your point on breadth thrusts being bullish for the IT/LT.



==============================================================
5. Sentiment: Extremes in sentiment tend to signal risk of a reversal is growing. The longer the extreme exists, the more likely a reversal will occur. Note: The best bet is to wait until sentiment peaks or troughs. Anticipating sentiment extremes is risky.
==============================================================

I generally agree, except that I have seen bits and pieces of anecdotal evidence suggesting that "Sentiment Thrusts" (to coin/borrow/butcher a phrase) may also be bullish IT/LT.


==============================================================
6. Monetary Conditions: Falling interest rates and an accommodative monetary policy tends to be bullish for stocks. Don't fight the Fed.
==============================================================

Again I agree, but... Keep in mind that the Fed cut rates many times before there was any noticeable effect. This leads me to believe that the tax cut package is having at least as much effect on the markets today as is Fed policy.


I hope this is helpful, or at least interesting. Barring that I hope is is at absolute minimum entertainingly funny, (albeit unintentionally).

(:

augie