Thursday, July 19, 2007

A Simple Measure of Trend and Trajectory

Of interest to traders is not only the trend of markets, but their trajectory. A rising market that is losing steam tells a different story from one that is picking up momentum.

I was reading a past article in Formula Research and came across a simple indicator that was used to assess relative strength. The indicator took the price changes for each market over 5, 15, 25, and 35 days and simply averaged those together. In so doing, they created a composite trend measure that is sensitive to trajectory.

I've charted this trajectory measure for the S&P 500 Index (SPY) going back to the start of 2004. (Click on chart for greater detail). You can see that the trajectory measure has tended to top out well ahead of price during market rises and that a dip toward or below the zero level has tended to precede final pushes to market highs. Trajectory has also tended to bottom coincidentally or ahead of price during recent market declines.

Trajectory has made a meaningful difference in terms of future returns. When the trajectory has been above zero (N = 581 trading days), the average gain in SPY 20 days later has been .35% (366 up, 215 down). When trajectory has been below zero (N = 272), the average 20-day gain in SPY has been a much stronger 1.50% (197 up, 75 down).

Interestingly, when we look at very positive trajectory markets vs. weaker (but still positive) ones, we also see a difference in future returns. When trajectory has been above 3% (N = 103), the next 20 days in SPY have averaged a respectable gain of .93% (80 up, 23 down). When trajectory has been above zero but below 3% (N = 478), the next 20 days in SPY have averaged a gain of only .23% (286 up, 192 down).

Finally, when trajectory has been very weak (below -3.0%; N = 37), the next 20 days in SPY have averaged a very strong gain of 2.89% (31 up, 6 down).

These results mirror those from my investigations of momentum with the Demand and Supply indicators. We tend to see above-average returns from very strong markets--those with high upside momentum/trajectory--and from very weak ones. Moderately strong markets have yielded subnormal returns.

There are many interesting applications of this research. For instance, along the lines of the Formula Research work, we can track the trajectories of different sectors to map out a trading strategy based on relative strength. We can also investigate trajectories and returns over longer time frames with weekly data or over shorter periods with intraday data. It's relatively easy to whip up a trajectory measure for any instrument--stock, futures contract, or index--with a price history. It also wouldn't be difficult to assess trajectory for market indicators, such as advance-decline lines (along the line of a McClellan Oscillator) or sentiment measures (options ratios; VIX).

I'll be posting more on this topic, exploring the applications.


Trajectory--Not Just Trend--Is Your Friend