Craig,
<<Herb, Sorry, I misunderstood your first post that price was one of 50+ factors. I meant
to ask if you had tested it on a market cap filter alone. Whatever the others are they
certainly seem to reinforce a strong price correlation; and compelling in both the high
and low ranges.>>
While the 6 month %returns I previously posted were based
on PRICE LEVEL alone, I use 50+ variables in ranking stocks.
The returns I get using 50+ variables are far better than those
shown for PRICE alone.
Happy to do some studies for MarketCap alone for a couple of 6
month holding periods. We know what they will look like, more
or less, given all the media bleating about big cap/small cap stocks.
<<After Bob let this code become part of the public domain Richard suggested the
refinement of using float factors on volatile / high ATR output as you noted. I've used
Mov(V,34,S)>50K vs 250K to include more thinly traded issues resulting in 1200, but
after the following float screen about 400 candidates remain - a reasonable group to
review with other tests for candidates>>
Looking at the scan you posted with output to ATR89.lst,
if it would be useful, I could do a study or 2 for stocks with:
a) PRICE of 5 & up
b) VOLUME of 250k & up
c) some measure of volatility like QP2 BETA. I don't retain the
. prior 89 days of Closes to do the kind of average volatility
. you use.
d) small float
What kind of holding period? |
