from the desk of Dr. Terry F Allen

Skip navigation

Member Login  |  Contact Us  |  Sign Up

1-800-803-4595

Testing A Statistical Rule of Thumb

Last week was the worst week for the market in over two months.  How did your stocks fare?  A week ago, I showed you the risk profile graph for our 10K Bear portfolio.  It spelled out what would happen at the various possible ending prices of SPY on Friday.  This portfolio gained 27.7% (after paying commissions) last week.  It has now gained 47% over the last 3 weeks while the stock has fallen 6%.

Our Boomer’s Revenge portfolio (which takes a neutral position on the market) enjoyed an expiration month where SPY fluctuated all over the place but ended up just about where it started.  That is the ideal thing for us.  This portfolio gained 45% in one month after commissions.

Where, besides options, can you make gains like this when the market is flat (or goes down)?  Most investors would be happy with 40% gains for a two-year period.  We did it in one month.

We take the position that we have no idea which way the market will move in the short run.  However, if we really could guess its direction, we could make extraordinary gains.  Today I want to check out one statistical rule of thumb that some people believe might give us an idea of short-term market direction.

Testing A Statistical Rule of Thumb

We have always held the position that we had no idea which way the stock was headed in the short run (but over time, it has gone up about 10% a year, so we should lean a bit in the upside direction).  This assumption seems to have served us well over the years. 

However, a time-tested rule of thumb in the world of statistics is that for almost any curve of economic activity, the direction of the change in the next period is more than twice as likely to be in the same direction as the direction of the change in the last period.  This rule of thumb supposedly applies to all time periods, whether they are daily, weekly, monthly, or annually.

For sure, if you examined a moving average curve for a stock or index, every subsequent change is almost always in the same direction as the last change.  Very rarely does the curve hit a bottom or top and reverse direction.

I wondered if we could use this rule of thumb to predict the direction the market might be headed based on what it did in the previous month.  If more than twice the time, the change was in the same direction, we could benefit considerably by setting up risk profile graphs which had more room for the stock to move in that particular direction.  If this statistical rule of thumb held true, maybe we should change our basic assumptions and the strategy as well.

To check out the idea with SPY, I checked the monthly changes in SPY over the last 10 years.  In exactly 60 of the months, the stock moved in the same direction that it had in the previous month.  In exactly 60 of the months, it moved in the opposite direction that it had in the previous month.   A perfect tie.

 The only conclusion that we could make from these numbers is that our original assumption that we just don’t know is probably about as good as we are likely to get.  The statisticians are not always right when it comes to the real world, at least when it comes to monthly fluctuations of SPY.

Tags: , , , , , ,

Making 36%

Making 36% — A Duffer's Guide to Breaking Par in the Market Every Year in Good Years and Bad

This book may not improve your golf game, but it might change your financial situation so that you will have more time for the greens and fairways (and sometimes the woods).

Learn why Dr. Allen believes that the 10K Strategy is less risky than owning stocks or mutual funds, and why it is especially appropriate for your IRA.

Order Now

Success Stories

I have been trading the equity markets with many different strategies for over 40 years. Terry Allen's strategies have been the most consistent money makers for me. I used them during the 2008 melt-down, to earn over 50% annualized return, while all my neighbors were crying about their losses.

~ John Collins