Wednesday, September 17, 2008

A Picture is Worth a Thousand Words

If there is anyone who reads these postings regularly, he, she, or it has probably noticed that I tend to pay more attention to the different versions of the Standard and Poor's stock reports (as opposed to Standard and Poor's Stock Reports) than I do to most analyst writings. A few days ago I was looking at the August 30, 2008, Stock Report for AA and I noticed something very strange. The report includes a chart of the stock price from January 2005 to the present with various overlays, including changes in S&P's own 12-month target price. In most of the cases where there was a large change in the 12-month target price - arbitrarily taken to be $4.00 or more, based on a casual glance at the relevant overlay there - the stock price six months later was also significantly changed, but in the opposite direction. Here are the numbers; much of the information, including the numbers for the target price, is only approximate, since it is taken directly from the graph, but even assuming a few small mistakes on my part, I believe that the pattern is sufficiently clear.
DateOld TargetNew TargetChangePrice thenNew DatePrice on N. D.Change
9/22/053329-425.903/21/0629.14+12.5%
3/8/062939+1028.789/7/0628.55-0.7%
9/11/063935-427.163/9/0733.20+22.2%
9/7/073640+434.873/6/0838.37+10.0%

A view of all of the changes in S&P STAR rating shows a similar pattern.
DateOld STARNew STARChangePrice thenNew DatePrice on N. DateChange
1/3/0643-129.906/30/0632.36+8.2%
6/5/0634+131.3812/4/0631.18-0.6%
2/7/0743-132.158/6/0635.67+10.9%
5/4/0732-135.6611/2/0838.56+8.1%
8/4/0723+131.042/1/0834.28+10.4%
1/22/0834+134.692/8/0833.76-2.8%
2/14/0832-135.368/13/0832.25-8.8%

Like any other numerical data in the investing world which are very regular, these price target data are impressive, and it seems likely that someone might find some practical use for the pattern. However, since the direction of the regularity is unexpected, it needs some explaining. My first guess at an explanation was that the phenomenon was related to the fact that the predictions were for 12 months, while I was measuring at six months; unfortunately, although the negative correlation at 12 months is less perfect, it still seems to be there.

My second guess was that either this analyst is special, or this stock is special. A glance at MON suggests that one or both of these guesses might be near the mark; while the magnitudes of the changes there are way off, at least the directions tend to be correct.

In any case, before taking price targets - or anything else the analysts say - too seriously, it's worth checking to see how the relevant analysts have done in the past.

0 Comments:

Post a Comment

<< Home

      

to my broker review site