A Stock Market Model - Notes
A Stock Market Model - Notes
A Stock Market Model - Notes
______________________________________________________________________________________________
iv. There seems to be no end of possibilities when it comes to manipulating an index or its
components in order to come up with a so—called objective opinion on a market's
direction.
v. We break down internal indicators into two distinct types:Trend—following and
Trend—sensitive
vi. Rounding out internal indicators is market breadth, an extremely useful tool in
determining the technical environment.
b. External indicators
i. External indicators include anything outside of price action: the monetary environment,
crowd psychology, valuation, and economic conditions.
ii. At NDR we tend to lump valuation and crowd psychology in the The Sentiment
Component.
iii. We also tend to include economic indicators that can influence markets (e.g., inflation)
in the The Monetary Component and sometimes, to a lesser extent, the sentiment
category.
iv. So basically we're looking at two sets of external indicators: monetary and sentiment.
15. To build a truly diverse, objective model requires finesse, diligence, a bit of artistic panache, and a
strong knowledge of macroeconomic relationships, ranging from company fundamentals to
macroeconomic fundamentals to Fed watching.
16. Ironically, computing power takes a back seat in building good models.
Fab Five
1. To create the Fab Five, we combined individual models covering our favorite areas of the market.
2. However, the model has only four components—The Sentiment Component, The Monetary
Component, Fab Five Combo Component, and the Tape Component—but the tape is given double
weight.
3. Figure 30.1 shows the past 10 years of the Fab Five model.
4. The mode boxes at the bottom summarize the results for the full date range as well as for the past 10
years.
5. The top clip has an additional series on it that represents an equity line based on going long the S&P
500 when the model is above the upper bracket, going into T–bills when the model is between the
brackets, and shorting the S&P 500 when the model is below the lower bracket. This equity line is
only for perspective; it's there to give a close–up view of how the model has behaved in the past. The
stats tell a story, but not the whole story.
4. We would never deviate too far from the Fab Five's basic verdict. If it's bullish we won't be bearish: we
might be mildly bullish, but never bearish.
5. Even though this model isn't designed for trading, traders can still use it to make decisions about
whether they should have a bullish or bearish bias. If the short side is your thing then you should be a
little more careful if the Fab Five is in bullish mode, and vice versa. At any rate, we still look to other
inputs when evaluating our market position, but always in the context of the Fab Five.
Tape Component
1. There are seven individual indicators in the tape component.
2. As with all the other Fab Five components, and in keeping with our KISS philosophy, these indicators
are each assigned—1, 0, or +1, and the result is produced through simple addition.
3. The first indicator for Tape Component is the Golden cross.
a. The first indicator in the tape component is shown in Figure 30.2.
c. The universe we are using for these calculations is the NDR Broad Market Equity Series
All–Cap Index. This index represents 99 percent of the market value of the common shares
trading on the big three exchanges. Its history goes back to September 1980.
d. The mode box in Figure 30.4 shows that when demand has been above supply, as it has been
79.0 percent of the time, the S&P 500 has advanced at 11.8 percent a year. But when supply
is above demand, the market retreats 1.2 percent per annum. This indicator's contribution to
the Fab Five tape component is +1 when demand is above supply and -1 otherwise.
6. Fourth and Fifth Indicator is to look for excitement in the stock market
a. There are many ways to look for excitement in the stock market: exuberance, whether rational
or irrational, can be measurable.
b. A couple of our favorite indicators look at the relationship between advancing and declining
volume, or advances versus declines.
c. Figure 30.5 evaluates the ratio of advances to declines. The ratio is actually a 10–day sum of
NDR Multi–Cap advancing issues divided by a 10–day sum of declining issues. The NDR
Multi–Cap universe covers the top 97percent of tradable market cap and includes the NDR
Large Cap plus the NDR Mid –Cap plus some of the NDR Small Cap (enough to get the
market cap to 97 percent).
d. When the advance/decline ratio exceeds 1.9, we call that a breadth thrust. The table in Figure
30.5 shows the history of this indicator back to 1947. Now I know what you're thinking: Hold
on—if the Multi–Cap started in 1980, just exactly how did you do this? Especially given that
lecture about following only common stocks on the NYSE? Well, we consider ourselves
market historians, and we love historical data—as much as we can get. So yes, we borrowed
the NYSE advance/decline data going back to 1947 to help illustrate the point that breadth
thrusts can be quite significant. It also turns out that, according to our database, stocks other
than common stocks made up only 20 percent of listed NYSE issues in 1980.
e. Now how exactly do we get these thrusts to work for us? Typically, we try to determine how
long initial momentum is likely to last based on history and then refine our prediction from
there. In the case of the indicator shown in Figure 30.5, 80 market days seemed to work well.
But what if a signal is repeated in that period? One could approach it in at least two ways:
either by simply resetting the count every time a new signal occurs in the 80–day window or by
ignoring that signal. After some quick testing we decided on the latter
f. The contribution of this indicator to the Fab Five tape component is either +1 or 0. It can never
go negative.
7. Sixth Indicator is a cool breadth gauge, the high–low logic indicator—developed by Norman Fosback
and shown in Figure 30.6—is it.
addition to the model's current level: when Big Mo Tape has been high and rising, the returns
have tended to be best, while a low and falling model reading has been associated with the
weakest returns
e. The weight of this indicator in the tape component of the Fab Five is tricky but sensible. When
Big Mo is rising and above 56 we assign a +1; when it's rising and below 56, it gets a 0; falling
and above 79 is another +1; falling and between the brackets is a 0, while falling and below 56
gets the only −1 reading. If Big Mo is unchanged on a six–week basis the reading from the
prior week is used.
Sentiment Summary
1. The final sentiment component, shown in Figure 30.20, is a sum of the indicators discussed (plus one
more P/E indicator) for a range of –8 to +8 (remember, the daily sentiment composite has double
weight).
2. The brackets are set at –1.5 and 2.5. Because the brackets are asymmetric, it's easier to get a sell
than it is to get a buy.
3. As with all of our other Fab Five components, this model was constructed to work in all four of our
preset time frames.
2. The brackets are set at –1.5 and 2.5, and to be consistent with the other major Fab Five components,
the model is designed to work in four preset date ranges.