|
|
|
|
 |
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. |
Write to us at:
(address is not clickable)
Q:
Larry Connors had interesting post on Tradingmarkets.com. He claims that his testing showed that if one had bought an index (SPX, NASDAQ, or SOX) on the open the day before last day of month and sold on open on 5th day of next month, only when prices above simple 200 day MA, the following gains would have been realized: (test period was from Jan 1995 to end of 2004) SPX 753 pts. vs. 743 pts. for buy and hold; NASDAQ 1768 pts. vs. 1217 pts. for B+H; and SOX 952 pts. vs. 293 pts. for B+H. He didn't mention any drawdown amounts you would have taken and figures might have been different trading SP500 and NAZ 100 futures. (I don't understand why futures contract data wasn't used). His claim is that one could have gotten outsized gains (for NAZ and SOX anyway) over B+H being in market less than 20% of time.
I would have liked to see a much longer test period, even on indexes, as this period had huge up bias in the late '90s and early 2000's. I would also have liked to see drawdown amounts. You would not have made any trades for long period of time indexes were under 200 day MA. Pretty boring. Still, premise is interesting and possibly worthy of further review. Any comments?
T.M.
Chair's response:
I agree. The tendency of money to come in at the end of the month is probably a regularity that rational expectations hasn't hit yet. Of course, most of the effect comes on the first day of the month, and all else is insignificant. The use of adventitious starting and stopping points is a typical bias of palookas. The upward drift makes it such that you could find almost as much for any 5 days randomly. Yale de la Mancha been talking about this spurious effect for 15 years, and we hope one didn't try it this January. Like most such published shibboleths, advice like this is always behind the form.