Hi Since Stephen asked why I included a paper that excluded stock market predictions from its analysis, here's the chain that led (erroneously) to my citing the Grove meta-analysis.
1. I remember reading a newspaper article some time ago about investment firms that were moving to actuarial predictions because they were outperforming analysts' judgments. 2. When the stock market came up on TIPs, I googled to see if I could recover something along the lines of stock market and actuarial versus clinical prediction. The closest thing that I found relatively quickly was the Grove paper. Its summary table did include a line for Financial and I merrily sent it on its way. Presumably, financial here meant something like personal financial decisions rather than the stock market. My apologies for shooting from the hip. I've done a bit more searching now and still nothing right on point that I know. Although the literature on prediction, Meehl, and the like does include provocative papers, including older ones, with titles like: Cowles, Alfred, "Can stock market forecasters forecast?" Econometrica, 1 (1933): 309- 324. I don't have access to the paper, but the following appears to demonstrate that people can actually perform quite well if they are ignorant enough of the stock market to benefit from a crude recognition heuristic; that is, invest in stocks with name recognition, a heuristic not readily available to experts. But the recognition effect failed to replicate in one subsequent paper by Patric & Rakow. Borges, B., Goldstein, D. G., Ortmann, A., & Gigerenzer, G. (1999). Can ignorance beat the stock market? In G. Gigerenzer, P. M. Todd, & the ABC Research Group, Simple heuristics that make us smart (pp. 59-72). New York: Oxford University Press. No more time for this now, but perhaps later. Take care Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax [EMAIL PROTECTED] Department of Psychology University of Winnipeg Winnipeg, Manitoba R3B 2E9 CANADA >>> <[EMAIL PROTECTED]> 24-Nov-08 8:46 PM >>> On 24 Nov 2008 at 2:15, Jim Clark wrote: > I'm not too certain what Stephen would include under "technical > analysis", but it does appear that the classic superiority of > actuarial over clinical prediction applies to the stock market as well > as numerous other domains. See: > > http://www.psych.umn.edu/faculty/grove/096clinicalversusmechanicalprediction.pdf > > I guess I'd better clear this up. I said "so-called "technical analysis"" which was meant to indicate that I was referring specifically to the loopy system of divination in the stock market based solely on the prior rise and fall of the stock. If someone was unfamiliar with this system, they might think I was referring to the actuarian approach, but I wasn't. Paul Bernhardt had it exactly right in his post (helpful urls too). What is truly scary is how many people in the investment business actually believe this stuff. No wonder we're in the trouble we're in. I also think the actuarian approach (fundamental analysis?) to stock prediction is doomed (see: theory of the efficient market) but at least it's an honest attempt. Yet I'm puzzled by Jim's citing the Grove et al (2000) study in support of its use for predicting the stock market. In their methods, they say "Only studies within the realm of psychology and medicine were included. Thus, we excluded attempts to predict nonhuman outcomes (e.g., horse races, weather, stock market prices)". Stephen ----------------------------------------------------------------- Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University e-mail: [EMAIL PROTECTED] 2600 College St. Sherbrooke QC J1M 1Z7 Canada Subscribe to discussion list (TIPS) for the teaching of psychology at http://flightline.highline.edu/sfrantz/tips/ ----------------------------------------------------------------------- --- To make changes to your subscription contact: Bill Southerly ([EMAIL PROTECTED]) --- To make changes to your subscription contact: Bill Southerly ([EMAIL PROTECTED])
