New Research Released…

Summary

The Dow Industrials will close higher at year end 53.4% of the time when January also ends the month higher.

The Dow will close lower 22.7% of the time when January also closes down.

Combined, the Dow follows January’s performance 76.1% of the time.

January and December are the best months to predict the following year’s up trend, with September being the worst predictor of an upward trend.

For downtrends, September is the best month with June coming in second.

You can download a spreadsheet of the results here (65k .xls zip file).

Background and Methodology

I used data on the Dow industrials going back to 1928 and programmed my computer to find the year end values for the Dow,
and the month’s end values. I compared how often the Dow closed higher after January closed higher, and how often the Dow
ended the year lower when January was also down.

Test 2

I also checked the other months to see if they were better predictors of performance.

For that test, I used a 12-month period, from January end to the next year’s January end as one sample period. Then Compared the February to February period, and so on.

In this way, each month received 12 months of performance. That differs from the prior test where the measurement period is from January end to December end (11 months).
If I used this approach for the other months, each succeeding month would have less time: February to December then March to December, and so on.

Results

You can download a spreadsheet of the results here (65k .xls zip file).

For the first test: As goes January so goes the year. That saying is true. Here’s what I found.

When you combine the matching upward trend (January up, Dow up, 53.4% match) with the downward trend (January down, Dow down, 22.7% match), you get 76.1%. In other words,
the Dow follows January’s direction 76% of the time.

If January closes up, there is a 53.4% chance that the year will also close up. If January closes lower than where it started, the year will also close lower but just 22.7% of the time.

The table shows the various combinations of up and down for January and the Dow industrials.

However, is January the best predictor of future performance or is there another month that works better?

Test 2

In this test, I used a 12-month period for each month. That is, January to January, February to February, …December to December. I used the last day of each month in the comparison as
explained in Background and Methodology.

Here’s what I found.

For example, I found that when January closed the month higher, the Dow also closed higher 12 months later (not 11, as in the previous test) 48% of the time.
September is the worst month at predicting the following upward close a year later, with just 26% of predictions correct. January and December tie for the best prediction rate: 48%.

Similarly, when January closed the month lower, the Dow also closed lower 14% of the time. The best performing month for downtrends is September. That’s no surprise because
September closes lower most often.

Combining the two rows we can says that the Dow follows January’s trend 62% of the time, the best of the bunch. November comes in second with 60%.

The worst prediction comes from September with just 48% seeing the Dow follow September’s direction 12 months later.

— Thomas Bulkowski

Written by and copyright © 2005-2017 by Thomas N. Bulkowski. All rights reserved. Disclaimer: You alone are responsible for your investment decisions.
See Privacy/Disclaimer for more information.
From Stephen Wright: I’m writing a book. I’ve got the page numbers done.