On Tue, 4 Apr 2006 14:29:06 -0700, David L Cassell <davidlcassell@MSN.COM> wrote:

>akramer@CERNER.COM wrote back: >>There are 75 predictions, each just a simple linear equation (i.e. the >>logistic model has already been run, it's just a matter of plugging in >>values for the independent variables). Each "equation" resides in a >>separate file that. > >Okay, first off, I would not keep each 'equation' in a separate file. >That is making your work harder. If all your equations were in a single >'driver' file, you could read that information into an array or a hash, and >then have that structure handy as you passed once through your >100,000 record data set, using the saved information to do computations >as you go. After all, if you have the formulas in a table held in RAM, you >can do all the processing as you read the record.

I agree.

I would also reconsider the structure of the variables which control which predictions are to be made for a particular case. I suspect that normalizing (one short row for each case/prediction pair, rather than one long row for each case) will make the computations simpler to implement.

> >I'm not even going to discuss the modeling issues here. > >HTH, >David >-- >David L. Cassell >mathematical statistician >Design Pathways >3115 NW Norwood Pl. >Corvallis OR 97330 > >_________________________________________________________________ >Express yourself instantly with MSN Messenger! Download today - it's FREE! >http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/