Daily Speculations The Web Site of  Victor Niederhoffer & Laurel Kenner Dedicated to the scientific method, free markets, deflating ballyhoo, creating value, and laughter;  a forum for us to use our meager abilities to make the world of specinvestments a better place.

Home

The Chairman
Victor Niederhoffer

1/13/05
Father Random, Part II

Father Random is so prevalent in the market that I hesitate ever to look at a seasonal, a serial correlation, a correlogram, or the accuracy of a turning point, the prediction of a reversion to a mean, or the consistency of performance. You see, things have a way of seeming regular, and the human mind is able to frame so many hypotheses, or fool itself with self-reported results so readily, even when the information or test is completely consistent with randomness, that one must always be skeptical. This would apply to things like the January effect. After all, the market goes up in 70% of all 11-month periods and there are only 30 or 40 years that would seem relevant to today, even if the cycles and Bacon were not ready to eat up those who rely on fixed rules or attention-getting, startling predictions a la Prechter and Roach.

However, it's conventional wisdom already that this January, down some 2%, is one of the worst beginnings of year on record. What does it signify? Nothing except a slight, presumably random, tendency to go up to the end of the month, and a very strong tendency to go up to the end of year. Let's consider every decline of at least 2% from year-end the previous year to the second Friday of January.

Year of  Amount of           Move from mid-Jan  Move from Jan month end    
Decline  Decline to mid-Jan  to end of Jan      to end of year
1990     - 4%                - 3%               + 3% 
1991     - 4%                + 5%               +20% 
1996     - 2%                + 4%               +20% 
1998     - 2%                + 3%               +20% 

Thus I note a strong tendency for the large declines in the first two weeks of the year to be bullish for rest of year. A Mann-Whitney test is in order here, but without patting myself on the back I can say that I've done so many of these that I know it's about 1 in 20 by chance. I attribute this to the tendency for the public to always get in on the wrong foot. I note that the self congratulations of the bears, who either have already put their positions on after a lackluster 2004, or who hope to scare the daylights out of the bulls for promotional value, are contributing mightily to the public's proper place in the firmament, i,e. of losing more than they have any right to lose. But I note also that the Specs try to preclude the tendency of the public and our readers to make a monkey of themselves as much as possible with the occasional hand study.

P.S. I note incidentally that during the '80s there was only one decline in the first two weeks of the year.

P.P.S. I also note that a hand study gives a thousand times more insight and information than a computer study, and infinitely more info than an idle scary conjecture.

Read Father Random, Part I

For more of Victor Niederhoffer's writings, click here>>>