Favorite Links

Tuesday, November 24, 2015

Mark Hulbert - A top-performing stock-timing model says the market is likely to rise

A top-performing stock-timing model says the market is likely to rise

By Mark Hulbert

Published: Nov 24, 2015 5:12 a.m. ET

CHAPEL HILL, N.C. (MarketWatch) — Odds are in stock investors’ favor between now and the end of next week.

I say this not because I believe in the so-called Santa Claus Rally, an urban myth that I tried to debunk a couple of weeks ago.

The basis of my confidence is the stock-market-timing model that has one of the best long-term records of any monitored by the Hulbert Financial Digest (HFD): the so-called Seasonality Timing System created in the 1970s by Norman Fosback. Fosback, who at the time was president of the Institute for Econometric Research, currently edits an investment-advisory service called Fosback’s Fund Forecaster.

Not only has Fosback’s model performed better than any other market-timing system the HFD has tracked over the past three decades, but it also is the simplest. It is entirely mechanical, based on nothing more than the calendar. The rest of this week and all of next should benefit from the convergence of the two seasonal patterns on which his model is based: strength around the turns of the month and before exchange holidays.

If it sounds too easy, consider this: According to the HFD’s calculations, a portfolio that invested in the Wilshire 5000 index on the minority of days deemed to be seasonally favorable and which otherwise earned the 90-day Treasury bill rate, produced a 10.5% annualized gain between 1982 (when the HFD commenced coverage) and the end of this past October.

A buy-and-hold strategy over the same period would have gained slightly more — 11.3%, or 0.8 percentage point more a year. But that’s an unfair comparison, since Fosback’s model calls for being invested in the market for less than half the time. And 0.8 percentage point a year is a small price to pay for cutting risk by more than half. 


Before you rush out and bet heavily on a short-term trade, however, bear in mind that there are no guarantees. Though the stock market has a well-above-average chance of rising over the next two weeks, it still could decline.

Since 1896, for example, when the Dow was created, the stock market has risen about two-thirds of the time over the period beginning two days before Thanksgiving and lasting until the end of December’s first week. That’s markedly better than the 54% odds of rising that prevail on all other comparable periods in the market, but it still means that stocks fall one out of three times over those pre- and post-Thanksgiving sessions.

If you really are serious about exploiting these seasonal patterns, you will need to bet on them consistently over many years. Your situation is not dissimilar to the odds a good blackjack player faces: By being a shrewd card counter, he can perhaps increase his chances of winning a given hand to 55%, from below 50% for a non-card-counter. Notice that this card counter’s odds of success when playing a single hand are still hardly better than a coin flip. However, by playing a sufficient number of hands, the odds of coming out a winner become closer to 100%.

Something similar may apply to the Seasonality Timing System. Your odds of success in any given instance may be statistically significant, but still not high enough to justify throwing caution to the wind. That’s why followers of the Seasonality Timing System need to follow it with discipline and patience over many years.

In any case, notice that there are good odds that our portfolios will be fatter in two weeks than they are today — just like our waistlines!

No comments:

Post a Comment