Friday, December 08, 2006

Assessing Market Psychology With Relative Options Indicators

In my last post, I proposed a different way to construct put and call options ratios. Specifically, the equity put ratio looks at equity put option volume as a fraction of its own 100-day moving average. Conversely, the equity call ratio measures equity call option volume as a proportion of its own 100-day moving average. The relative put/call ratio is the equity put ratio divided by the equity call ratio.

Going back to 2004 (N = 720 trading days), we have had 54 occasions in which the five-day equity put ratio has been above 1.2 *and* the five-day relative put/call ratio has been above 1.2. What this means is that we've had a five-day period of higher than average put activity relative to call activity that can be attributed specifically to speculator interest in puts.

Ten days following this spike in put activity, the S&P 500 Index (SPY) was up by an average of 1.11% (42 up, 12 down)--quite a bullish edge. The remaining days in the 2004-2006 sample were up by an average of only .25% (391 up, 275 down). It thus appears that, when put activity is high relative to call activity and relative to its own recent level of activity, returns are above average.

Now let's look at occasions in which the five-day equity call ratio has been above 1.2 (20+% more call options traded relative to the 100-day average) *and* the five-day relative put/call ratio has been below .90 (N = 69). What this means is that we've had a five-day period of higher than average call activity relative to put activity that can be attributed specifically to speculator interest in calls.

Ten days following this spike in call activity, SPY was up by an average of .69% (46 up, 23 down). That is considerably stronger than the average ten-day gain for the rest of the sample (.28%; 387 up, 264 down). It thus appears that, when call activity is high relative to put activity and relative to its own recent level of activity, we see superior returns over a two-week period.

What I'm seeing in the data so far is evidence of reversal effects (for puts) and momentum effects (for calls). When we get a significant expansion of put activity, it's usually in a falling market and that falling market tends to reverse itself. When we get a significant expansion of call activity, it's usually in a rising market and that bullishness enables the market to sustain its rise. In the current market, we see neither strongly elevated put or call activity on a five-day basis.

Clearly, there is much more to be studied in the options data, particularly at an intraday level. So far, at least, it does appear that the sentiment of options participants does have an impact on the distribution of future price changes in the stock indices. The way to superior returns, I suspect, is to look at new market data or to explore old data in new ways. Trading success, in that context, requires an unusual combination of tight discipline and open-minded creativity.