Kevin_in_GA
4,599 posts
msg #104553
- Ignore Kevin_in_GA |
1/20/2012 1:39:09 PM
NKTR taking a nosedive today. Since there were only two positions this week, the week will be a net loss. This happens in every system from time to time. My backtesting in Stratasearch is consistently returning a smaller set of stocks matching these criteria than what SF returns. Not sure why, but this will make the backtests different (although hopefully not really all that different in terms of overall performance).
I recoded this for a short filter as well - Mahkoh was right in his statement that it will work both ways. It just took me a little while to validate and backtest the code.
I think this will be a very powerful filter set (both long and short). I am going to move this to a new thread with the correct validated code and backtest data for both gap up and gap down filters.
|
mahkoh
1,065 posts
msg #104561
- Ignore mahkoh
modified |
1/20/2012 5:51:33 PM
Running the filter and adding the syntax: " open1 below horizontal" reveals that there was another trade this week.
ELGX filled a nearly 3 % gap on Wednesday after having moved lower on the first trading day of the week.
Must have missed that as its avg volume is also below 400K. If we include that trade the long version breaks about even for this week.
Running the short version and adding "open1 above horizontal" has me puzzled: The top 5 does not include SWC MAKO, FUL, or KEG. As a matter of fact FUL shouldn't even have qualified as it shows gap:gapfill ratio of 5:7. Subtract this week's gap and fill and you are left with 4:6 which puts it at 66% gapfill.
Actually the "true" short version underperformed this week with MELA, MNKD and QCOR filling their gaps but suffering a big loss from HOGS and UEC. Which makes another case for applying a stoploss in the short version.
|