TRADING TECHNIQUES

Statistically Analyzing Volume


by William M. Goodman, Ph.D.

Is an increase in the activity of a stock a meaningful indication of the direction of price? To answer this question, you can use statistical methods and most any spreadsheet to analyze the relationship between price and volume.

"How can you be sure a new technical indicator has merit? In the case of a price pattern, you test the pattern over different time periods or sets of charts, making buys and sells at the points suggested by the technique. This simple research can easily answer the pertinent question: Would trading based on the price pattern make you money?"

"Let's look at an example of a more elaborate model of price behavior. Figure 1 begins a test of the oft-claimed relationship between stock trading volumes and price changes. The relationship is the magnitude of the price change relative to normalized volume. The full test is based on a random sample of data for 50 stocks during different periods. Figure 1 illustrates what I refer to as the classic 80/20 pattern: Most of the effects you care about, the 80%, are caused by or happen to a minority of the samples, the 20%. In this case, virtually all the largest price surges occur during a few days when the volume hits the highest category."

Figure 1: TRADING VOLUMES. For all trading volumes, daily price changes below 5% are most prevalent. But at higher volumes, there is a move toward larger price surges.

Confirmation bias can occur by 
focusing too much on confirming evidence while possibly ignoring or undervaluing disconfirming evidence. With so many stocks and exchanges to track, you may unconsciously zoom to the more promising charts.

"Our task is to validate claims being made about the real world, based on data from a selection. The irony, if you hope to prove that a real or nonrandom effect exists, is that you must deliberately seek randomness at the start."
Excerpted from an article originally published in the November 1996 issue of Technical Analysis of STOCKS & COMMODITIES magazine. 
© Copyright 1996, Technical Analysis, Inc. All rights reserved.

Return to Back Issue Archive or November Contents