No plausible theory has yet been advanced for why stocks should be more volatile in October than in other months

October is indeed the most volatile month for the stock market, but that's mostly because of just three years: 1929, 1987 and 2008.

Although October historically has been the most volatile month of the year for stocks, you shouldn't bet that it will continue to be in the future.

That's because there's no plausible theory to explain why it should be more volatile than any other month. And without such a theory, odds are very high that October's past volatility is a statistical fluke. And that, in turn, means there's no reason we should be especially skittish in October when placing stock-market bets.

That doesn't mean there aren't other reasons to be nervous. But if you are getting anxious just because the calendar is about to turn to October, you can relax.

The accompanying chart plots the Dow Jones Industrial Average's DJIA volatility for each month since the benchmark's creation in 1896 (as measured by the standard deviation of daily percentage changes). Notice from the green columns that October is indeed the most volatile month, with a standard deviation of daily changes of 1.41%, versus an all-month average of 1.10%.

Notice also, however, that this result is heavily dependent on just three years: 1929, 1987 and 2008. The Octobers of those three years, of course, include the two with the worst crashes in U.S. history as well as one when the bottom dropped out of the market during the global financial crisis. Without those years, the standard deviation of October's daily changes is only marginally higher than the all-month average (1.18% versus 1.05%), and almost identical to that of March (1.18% versus 1.16%).

Statisticians I have consulted insist that it's almost certainly a coincidence that those three particularly terrible months for the market happened to occur in October. If the only historical data investors focused on were from the other 127 of the past 130 years, it's highly doubtful that October's volatility would be thought to be noteworthy.

Still, investors have a hard time giving up on old wives' tales, and many will continue to point to October being the most volatile month (if only marginally) even after excluding 1929, 1987 and 2008. It's important to point out to these skeptics that none of the explanations that have been offered for October's volatility can withstand historical scrutiny.

Here's a list of a some of the more popular explanations:

-- October's volatility can be traced to it being the month immediately preceding midterm and presidential elections. This explanation has some potential validity, since it makes sense that there would be heightened uncertainty prior to elections. But this uncertainty doesn't show up in the stock-market data: October's volatility is actually higher in non-election years than it is prior to midterms and presidential elections.

-- October's volatility can be traced to the end of mutual funds' taxable years on Oct. 31. Once again, this explanation has some superficial validity, since mutual funds, just prior to their fiscal year-end, have an incentive to sell their losing positions to offset any capital gains they may have realized over the prior 12 months. But this can't explain October's volatility, since the month was also the most volatile prior to 1986, which is when Congress passed a law mandating that all mutual funds use Oct. 31 as their fiscal year's end.

-- October's volatility can be traced to it being the first month of the third-quarter earnings season. For this explanation to withstand statistical scrutiny, then volatility should also be elevated in January, April and July - the first months of the other three quarters' earnings seasons. But that is not the case. Volatility in those other three months is actually well below average.

-- October's volatility traces to economic uncertainty being higher in that month than in other months. This argument is plausible, since uncertainty translates to volatility. But there's no evidence for it. In fact, the Economic Policy Uncertainty index in the U.S. actually is lower in October, on average, than it is for the average of all other months.

Of course, there may be some other theory for October's volatility that can withstand historical and statistical scrutiny. Absent such a theory, however, Mark Twain had it right when he famously wrote: "October. This is one of the peculiarly dangerous months to speculate in stocks. The others are July, January, September, April, November, May, March, June, December, August, and February."

Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at mark@hulbertratings.com

-Mark Hulbert