Move over, January: At least two more months are more predictive of stock returns than you are.
January has a reputation for being able to forecast the direction of the US market over the next 11 months of the year. This presumed ability is known as “January Predictor” and “January Barometer”.
You will see a lot of references to this indicator in the coming days, now that January is officially on the record books as a “down” month – with the S&P 500 SPX,
fall of 1.1%. I have already written that the January Predictor rests on a shaky statistical basis. Financial headlines will nonetheless announce the supposedly negative implications of January’s drop for the remainder of 2021.
So let me point out a few other ways the Predictor isn’t worth tracking.
What’s so special about January?
A good place to start is to remember that January 2020 was also a month on the decline (down 0.2%) and yet the following 11 months produced a gain well above the average of 18.4% (assuming that dividends have been reinvested).
It’s just a data point. Another clue that there is nothing particularly special about January is that other months have even greater predictive “powers” in forecasting the direction of the stock market over the next 11 months. Since the inception of the S&P 500 in 1954, in fact, June has the strongest forecasting capacity, followed by February. January is in third place.
Why, then, don’t you read about a June predictor or a February barometer? My hunch is that adherents are motivated less by statistical rigor than by stories and narratives that capture their attention. From a behavioral standpoint, the calendar year is a more natural time to focus on than the February to February or June to June periods. But the psychological significance is different from the statistical significance.
The importance of real-time testing
There’s another telltale sign that the January indicator isn’t all it’s supposed to be – it doesn’t pass real-time tests.
By that I mean testing done after it was initially “discovered”. If the January Predictor had passed these tests, we would be much more confident that this is not simply the result of a data mining exercise in which historical data is tortured long enough to cause a pattern to emerge.
But he couldn’t. As far as I know, the real-time test of the January Predictor begins in 1973. This is the first mention of it on Wall Street, according to an academic study on the subject. Unfortunately, his record since then has been much less impressive. Since 1973, in fact, not only is it not significant at the 95% confidence level that statisticians often use to determine whether a model is authentic, it is not even significant at the 85% level.
We shouldn’t be surprised; in fact, the January Predictor is in good company. Take the example of a study published last May in Review of Financial Studies. He looked at 452 supposed statistical models (or “anomalies”) that previous academic research had found. The authors of this recent study could not reproduce these results in 82% of cases. The remaining 18% turned out to be much lower than originally advertised.
No correlation between magnitude of rise and return to January in next 11 months
Another clue that January’s predictor has a shaky statistical base is that there is no correlation between market strength in January and its gain over the following 11 months. If there was such a correlation, perhaps we could whip up a plausible story about investor confidence at the start of the year that will continue for the rest of the year.
But there is no such correlation. Because of this absence, to believe in the effectiveness of the January predictor, one would have to believe that a gain of only 0.01 on the S&P 500 has as much predictive power as a gain of 13.2%. It tires credulity.
By the way, I chose this 13.2% in my illustration because it is the biggest January gain for the S&P 500 since its inception in the mid-1950s. It happened in 1987. From January 31st from that year to the end of 1987, the S&P 500 lost 9.9%.
To benefit from a statistical pattern, you must follow it religiously for years
Finally, even if the January Predictor had a solid statistical foundation, it would have to be acted upon for many years in a row in order to rationally attempt to profit from it. A good rule of thumb in statistics is that you need a sample of at least 30 before patterns become meaningful. In the case of the January Predictor, that means you’ll have to follow it for three decades. Also, during those 30 years, you would not make any other transaction other than switching to a 100% equity allocation every January 31 when the stock market rises in January, and to a 0% allocation if the market in January. is falling.
Without that patience and discipline, you don’t do much to improve your odds above those of a toss.
The bottom line? For all intents and purposes, you can’t conclude anything from the stock market’s decline in January about its position as of December 31st.
Mark Hulbert is a regular contributor to MarketWatch. Its Hulbert Ratings tracks investment bulletins that pay a fixed fee to be audited. He can be contacted at [email protected]
Following: Prepare for 10% drop in stocks, driven by 3 Rs, warns Bank of America
Read also : This Surprising Stock Market Synchronization Strategy Has Academic Seal of Approval