Hi lin luo; Maybe I didn't explain regression towards mean clearly enough.
Here are the revenue figures for MSFT for the last nine consecutive years, in billions of dollars:
Year Rev. %gain
1990 1183 ---
1991 1843 56%
1992 2759 50
1993 3753 36
1994 4649 24
1995 5937 28
1996 8671 46
1997 11358 31
1998 14484 28
See, for example:
www2.edgar-online.com
I don't have figures at hand for the earlier years of MSFT, but I believe it grew at above 50% for most of the 80s. The average growth for the last three years is 35%. For the first three years of the above, the growth rate was 47%. The sequence is clearly regressing towards mean, and has been doing so for a couple decades. As far as statistical analysis of stocks, this is probably the best example of regression towards mean that could possibly exist.
I am not trying to say that MSFT is a bad investment, or that they are now an average company, just noting that their growth is slowing. This is a classic example of regression towards mean. MSFT's growth is still above average, but no longer as far above average as it once was.
Incidentally, given exponential growth with a limit (i.e. a saturation point), there is a simple differential equation that can be fitted to company growth. There is a great book on the subject, complete with beautiful charts showing, for instance, the take-over of the transportation market by the car from the horse. Fitting those equations to MSFT suggest that it has about filled up half its niche. Given that, it appears to be overvalued at the current price to me.
I've been so busy engineering that I haven't had time to (insert vulgar description of disgusting, but necessary physical activity here) for the last two months, but I am looking forward to get back trading again. (Note that I seem to trade best when I hold for 10 minutes or less, so none of what I have stated about DELL or MSFT's future growth rates affects my trading in any way.)
Best of luck to you long term types...
-- Carl |
