Move over, January: at least another two months has more forecasting power for stock market yield than you have.
January has a reputation for being able to predict U.S. market direction over the 11 months of a year. This capability is called “January Predictor” and “January Barometer”.
You’ll see a lot of references to this brand in the coming days, now that January is officially in the record books as a “down” month – with the S&P 500 SPX,
slipping 1.1%. I have previously written that the January Predictor lies on a shaky statistical basis. But financial headlines trumpet the alleged effects of the January recession for the rest of 2021.
So let me point out a few other ways that the Predator is not worth following.
What is so special about January?
A good place to start is to remember that January 2020 was also a month down (down 0.2%) but nonetheless, the next 11 months delivered a well above the 18.4% average (a assuming that shares have been deposited).
That’s just one data point. One other view that is not particularly specific about January is that “powers” are even more predictable at other months when the stock market direction is predicted over the next 11 months. Since the creation of the S&P 500 in 1954, of course, June has the strongest forecasting capability, followed by February. January is in third place.
So why not read about Predictor in June, or Barometer in February? It’s my opinion that fans are less motivated by statistical strength than stories and narratives that capture their attention. From a behavioral point of view, the calendar year is a more natural time to focus on than the February-to-February or June-to-June periods. But there is a different psychological meaning than a statistical one.
The importance of real-time testing
There’s another story sign that the January Signal isn’t everything: It doesn’t go past real-time tests.
By this I mean tests that were performed after it was “first discovered. If January Predictor could do these tests, we would be much more confident that it is not just the result of a data mining exercise in which the historical data has been tortured long enough for a pattern to emerge.
But it was not possible. As far as I can tell, the January real-time trial at Predictor begins in 1973. That’s the earliest reference to it on Wall Street, according to an academic study on the subject. Unfortunately, the record since then isn’t as spectacular. Since 1973, of course, not only does it not matter at the 95% confidence level that statisticians often use when determining whether a pattern is true, it is not even important at the rate 85%.
We should not be surprised; in fact, January Predictor is in good company. Consider an audit that appeared in May last year in the Review of Financial Audits. It examined 452 apparent statistical patterns (or “irregularities”) of previous academic research. The authors of the recent study were unable to replicate these results in 82% of cases. The remaining 18% were much weaker than originally reported.
There was no association between the magnitude of January increase and return over the next 11 months
One other view that the January Predictor rests on a grim statistical basis is that there is no correlation between the market strength in January and the gain over the next 11 months. If there were such a correlation, we might be able to make a plausible statement about investor confidence at the beginning of the year carrying over for the rest of the year.
But there is no such correlation. Because of that absence, to believe in the effectiveness of January Predictor you have to believe that an S&P 500 gain of just 0.01 carries the same amount of prediction power as a 13.2% gain. That strains credit.
Anyway, I chose this 13.2% in my picture because this is the biggest gain in January for the S&P 500 since it was formed in the mid-1950s. That came in 1987. From 31 January of that year through the end of 1987, the S&P 500 lost 9.9%.
To profit from a statistical pattern, you have to follow it religiously for years
Finally, even if the January Predictor would rest on a solid statistical footing, you would have to work on it for several years in a row to go rationally trying to make a profit from it. A good rule of thumb is that you need a sample of at least 30 before patterns become meaningful. In the case of the January Predictor, this means you have to follow it for three decades. In addition, during those 30 years you would do nothing but move into a 100% equity allocation on January 31 in which the stock market will rise in January, and into a 0% allocation. if the market in January is down.
Without that patience and control, you are not doing much to improve your numbers above a dog whip.
The bottom line? For all intents and purposes, you can decide nothing from the January decline of the stock market about where it will stand on December 31st.
Mark Hulbert contributes regularly to MarketWatch. His Hulbert Ratings monitors investment newsletters that pay a flat fee for being audited. It can be reached at [email protected]
More: Get ready for a 10% stock fall, led by the 3 ‘Rs,’ warns Bank of America
Read also: This amazing strategy has the seal of academic approval for its stock market era