Ken:
One attracted my attention as it was written by an old Prodigy friend, Bill Sandusky. I just ran that scan (BNS)this morning and it ran about 31 minutes. Is that normal?
No, this isn't anywhere near normal. The reason this scan takes so long is that it has a divide-by-zero fix in it for Stochastics -- a fix added later. There are other versions of this scan without the correction, but the one with the correction is necessary if you're scanning the entire database, I think. That's because only a few stocks cause the error (and on some nights, there was no error, depending on the quotes for the day.) I have other versions of the BNS scan, which I'll try to get around to posting (one may be on the Website now -- the one without "zero error fix"). This is one of the few scans that required this divide-by-zero fix. Perhaps another fix could be devised -- one that doesn't slow down the works. For all I know, the inherent problem in the QP Stochastics formula that created the divide-by-zero problem may have been fixed. The scan was written quite a while ago. And without the fix, it hums along rapidly.
Try some of the other scans, such as Richard Estes' breakout scan (a terrific scan), which should run in much less time (around 3 minutes on a 233 mhz machine).
Brooke |
