Wednesday, August 16, 2006

Opening Gaps in the S&P 500 Index - Part Two


Note: To make it easier to automatically get updates from this blog, I've followed the suggestion of a couple of readers and added a FeedBurner link on the right side of the page. Click on that and you can add TraderFeed to your favorite browser or aggregator. This is an anonymous process and will not subject you to spam/solicitations. Eventually, TraderFeed will get to the point of offering intraday market postings and having automatic updating will be helpful during the trading day.



In the previous post, we saw that gaps between yesterday's close and today's open vary in size with the volatility of the market. In this post, we'll look at what I call the "relative gap". This is computed by taking the price change of the opening gap and expressing it as a proportion of the daily high-low range averaged over the past 40 days. All computations use SPY data.

Going back to March, 1996 (N = 2608 trading days), we find 197 occasions in which the opening range is .50% or more of the average daily trading range. That day's average price change from open to close is .09% (106 up, 91 down), stronger than the average daily open to close change of -.02% (1303 up, 1305 down). This strength tended to persist to the close of the following session, with an average price change of .14% (111 up, 86 down). That is stronger than the average price change from open to following day's close of .01% (1352 up, 1256 down).

During the period since March, 1996, we had 123 occasions in which there was a downside gap in excess of .50% of the average daily trading range. From the open to the day's close, the average price change in SPY was .43% (73 up, 50 down), very much stronger than the average daily open to close change as noted above. That strength also persisted into the close of the following day, with an average change of .66% (70 up, 53 down).

Interestingly, large opening gaps are associated with larger daily trading ranges. When the relative gap was .50% or greater, the day's average trading range was .92%. That is wider than the average trading range of .77% for the sample overall. When the relative gap was -.50% or weaker, the average trading range was a large 1.22%. It thus appears that large relative gaps have bullish implications--especially when they're to the downside--and that large relative gaps tend to be associated with more volatile action during the day trading session.