Article - Analysis of Daily Stock Price Volatility

2006-05-09

Stock price volatility is one of the most important aspects of studying the market. This article starts a series that will examine theories and assumptions about price volatility.

This article focuses on looking back through years of market data to determine if there are any general trends in daily price volatility. It's a commonly held belief that the open and close of trading are the most volatile times of the day. We'll determine if that's true and if so how much of an impact it has.

Looking for extremes

The first part of the analysis was to see if the hour of the day had an impact on the likelihood of a stock to hit an extreme. In this case we decided to look for price minimums. Below is a chart to represent the results. We measured the occurrences of a symbol having it's daily price minimum within the specified hours of the day on several randomly selected days. The count of occurrences on the left is not necessarily important, but the scale of them is.

The results are intuitive but the degree of difference is somewhat surprising. It wasn't surprising that the open and close are the most volatile, but the observation that the first hour of trading had twice as many price minimums as the 12th was. But, does this data really reveal a trend? The next logical question is whether there's a difference in behavior here for stocks that are increasing or decreasing for the day.

The above chart actually has some interesting implications. For a stock that's heading down on a given day the normal expectation would be for it's price minimum to occur later in the day, but here we have that dip in the middle of the day again. The behavior of the stocks heading up isn't surprising although intuitively the line should be heading further down at the end of the day.

General price volatility

The next 2 charts take a look at the average price volatility by hour. The data is based on the absolute value of percentage difference of the price from the open or previous close value. The data here is based on an entire month. Is there any correlation between hour of the day and volatility?

The chart above, of average % difference, has no surprises for us. Prices will diverge further from their previous close and open as more hours pass in the day. The only way this may not be true is if the market was truly moving sideways. It confirms the finding above about the middle of the day being less likely to generate a price minimum if we assume that extends to general volatility as well. Notice in this graph how the slope of the line in the middle of the day is lowest. That means prices are changing at the slowest rate at those points.

The graph above is a bit less intuitive than the others. The high volatility at open makes sense but it's a bit surprising to see the volatility not peak again at closing.

Overall there is definitely a correlation between hour of the day and prices changes/volatility, but one that makes sense.

What are the implications?

If you're trying to dump a stock you may have a slightly better chance of not hitting the daily minimum by trading in the middle of the day, over the long run. We did integrate a change to the automated back-testing strategy to prefer selling in the middle of the day, and saw a slight improvement, but nothing that turned a losing strategy into a winner.

Feedback

Please send us any feedback or requests for other data analysis to feedback.