8 Maiket Models

8.2 Sentiment Model

8.2 Sentiment Model

In 1986, Zweig devaloped a "Super Model", combining several monetary and momentum indicators to predict maiket direction [39]. Here, we re\iew seven different market indicators and then incorporate them into the Pattem Trading Sjstem (Chapter 3). By encoding the behavior of each maiket indicator, we can construct the Sentiment Model to sjnthesize the buhish and bearish behavior of each indicator and generate signals to predict market direction, emulating the Acme M sjstem.

8.2.1 Volatility Index (VIX)

The VolaliliiyIndex (AlX) measures the imphed volatilitj ofinde.x options. The AlX is huersely related to market direction; consequently, a high reading is associated with shaip correchons, while a low relative VIX reading marks the end of an uptrend.

Tcgefher on a chart, a broad-based market inde.x and the VIX wih appear as mirror images of each other (see Figures 8.2 and 8.3). Histoiicehy, high AlX readings can reach fifty and abo-e, whhe low readings bottom in the twenties. The beha™r ofthe AlX is asjmmetrical because as the VIX spikes up during maiket corrections, it declines graduaUy during maiket ad\ances.

SSPX LAST-Daily 12/D7/2QQ1

-iKiiroK.Z. V0I..1.I.IV lod.v (VIX)

Since we a\oid absolute values ofthe AlX. we look for new Mghs or lows in the form ofspikes o-er a reference period. Technically, aspike is a combination ofa channel breakout with a large range bar. Ifthe AlX spikes up, then a Buy signal wril be generated. Ifthe AlX spikes down, then a Sell signal wih be generated.

Fure 8.3. \TX Mirror Image

With this technique, we can identify extreme readings in the AlX and see spikes on the chart in either direction. Although this technique is good for identification, the trading signals have not been cleady defined. First, as with trading sjstem, we do not want to enter a trade without coitfiimaiion. Second, we do not want to restrict signals to spikes alone. As soon as the AlX makes a new high or low o-er a given range ofbars, we want to prepare for a confirmation.

Whai do we mean by confirmation? In the case ofthe VIX. as the market goes down and the AlX spikes up, we want to see the AlX first tick down before going long the maiket. This down tick in the is usuahy accompanied by an uptick in the market, since the two are inwrsaly related. Cleariy, we want to use this relationship as ageneral confirmation technique that can be applied to any indicator. The only question is whether an indicator is positively correlatedwifh the market (indicator rises as the market rises) or negatively correlated with the market (indicator falls as the market rises, e.g, the VIX).

There aretwotypesofconfirmation:/M/j cjhHih.iou and Ioh conjirmation. If the previous price is the hiesl price of given , I the ciffreiil price is I s t n the previ s pri , t n huti ifim irs If the prevu s

8 Maiket Models

price is the lowest price of a given range, but the current price is higher then the previous price, then a low confirmation occuis. The interpretation of a high or low confirmation depends on whether or not the indicator is positively or nega-tiveiy correlated with the maiket. Table 8.7 shows the signal to take based on the indicators confirmation and its market correlation:

Table8.7.1ndicatorConfirmation

From this confirmation Icgic, we created a function cdiedAcmeHigliLowIndex to test for both high and low confirmations. The Acme Maiket Sjstem calls the AcmeHigliLowIndex function separately for each indicator in the model. Each time, the function retums one ofthe following values to the Maiket Sjstem:

a 0 = No Confirmation

p 1 = High Confirmation

a 2 = Low Confirmation

The Market System then populates its long and short pattem strings based on the confirmation values The EasjLanguage code for the AcmeHigliLowIndex function is shown below in Example 8.1

Example 8.1. Function AcmeHighLowJndex

AcmeHlBhLowIndex: Calculate the High Low Index

Inputs:

Price(Nunieric), Length(Numeric);

AcnieHighLowIndex = 0;

If Price[l] >= Highest(Price, Length - l)[2] and Price < Price[l] Then

AcnieHighLowIndex -- l Else If Price[l] <= Lowest(Price, Length - i)[2] and Price > Price[l] Then

AfinrHiKhlowlridi-x ?;

8.2 SeritUHeritHidel

82.2 Put/Call Ratio

The Put/Cell ratio is an inde.x calculated by the Chicago Board Options Exchange ( ). The ratio compares the total put volume with the total call volume for stock options or index options. Investors buy more calls then puts, so the ratio never reaches one unless the market is declining sha ly. The put/cell ratio is negatively correlated with the maiket because people tend to buy puts at bottoms, and historicelly this behavior has pro\en to be wrong, as shown by the circled area in Figure 8.4.

Many traders look at the value ofthe put/cell ratio on a historical lereL For e.xample, a ratio greater then 0.8 is considered to be bullish, and a ratio less than 0.4 is considered be atish; however, we do not care about the absolute readings because we are using the confirmation technique.

Figure 8.4. Put/Call Ratio Peak

Figure 8.5 shows a trough in the puicell ratio. Although spikes up are common, spikes down are rare because such alow reading means that everyone is bullish. This situation is akin to everyone running to one side ofthe Titanic. Essentially, the whole countrj was long in March 2000, and nobody was left to buy. The charts illustrate how the put/call ratio is not entirely symmetrical. The fear of losing money is much more powerful than the satisfaction in making money, and the emotional trader usually makes the wron decision at the wron time.

SSPXLASTVeekly

Figure 8.5. Put/Call Ratio Trough

8.2.3 New Highs

Each day, the niunber ofstocks makhig 52-week highs on the New York Stock Exchange is tracked as the "NYSE New Highs" number The New Highs indicator is positively correlated with the market. As the maiket goes up, so does the number of new highs; however, the behavior offhe New Highs data is slightly different than the behavior ofits corresponding market index. When the maiket attains a new peak, the number of new highs spikes at the peak. As the maiket puhs back, the number of new highs drops close to zero. At the next peak, the new highs wih spike once again.

The key to interpreting new high data is to compare the new highs at two maiket peaks. Iffhe market is higher at the second peak, but the number ofnew highs is lower at its second peak, then a div-ergence has been created, as shown in Figure 8.6. The divergence in this example occurred just before a20o selloffin the S&P 500 mJuly 1998.

Fosback created an indicator in 1979 cahed the High Low Lcc Index in order to reccgnize these div-ergences [14]. The index calculates the minimum of two ratios: the ratio ofnew highs to the total number of issues, and the ratio of new lows to total issues. When the index is high, the market attgins both ahigh number ofnew highs and new lows, a bearish indication because market breadth is narrowing. When the index is low, then either the number of new highs or new lows is low, a bullish indication in both cases.

[BPXLASTJWeekly 0 /11/19

Figure 8.6. New Highs

8.2.4 New Lows

The number of stocks making 52-week lows on the NYSE is tracked as the "NYSE New Lows" number. The New Lows indicator is negatively correlated with the market, meaning that spikes in the number ofnew lows is a bullish indication, as shown by the circled areas in Figure 8.7.

ISPXLAST.Weelily 01/21/2000

* j