SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Strategies & Market Trends : MDA - Market Direction Analysis
SPY 681.92-0.7%Dec 31 4:00 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Psycho-Social who wrote (83836)5/18/2003 9:21:29 PM
From: Night Trader  Read Replies (1) of 99985
 
I dug out from my files a paper written by Ken Fisher and Meir Statman in the Financial Analyst’s Journal that compares sentiment surveys and next month returns from July 87 to July 98.

For the AAII poll they found an R squared of 5% for large cap and 1% for small cap. From eyeballing the scatter plot it would seem that any negative correlation comes from very high values of the survey – 50% or more bulls – and can be ignored otherwise.

For the II survey they found R-squared to be zero for both large and small cap, apparently confirming an earlier study that found no predictive value. I know Mark Hulbert believes in it (well he would wouldn’t he?) and I had a email correspondence with him over it but the evidence, at least over the short term, just isn’t there. The slope of the regression line was negative but not significantly so (in a statistical sense). Some newsletter writers beat the market and some don’t but taken in aggregate no conclusion can be drawn.

For WS strategists R-Squared was 3% for large cap and 1% for small cap. Unlike the other two surveys this was largely unaffected by recent returns i.e. it didn’t exhibit sheep-like momentum behavior.

Given the problems you’ve already stated and the recent popularity of these measures – particularly the II poll, which seems to be quoted by everyone – I really don’t see their usefulness, except possibly very high values of the AAII and WS strategists.

The II poll does serve a valuable function however; whenever you hear anyone quoting it approvingly, you know they don’t know what they’re talking about.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext