[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: ...stepwise regression...
At 11:48 13/04/2004, Christer.Thrane@hil.no wrote:
I totally agree with the most commonly expressed views on stepwise
regression on this list. But for the purpose,
Prior research and/or theoretical rationales and/or common wisdom are three
much used sources for selecting independent variables in a regeression
Remember Leamer's labelling of stepwise regression-- a.k.a. unwise
depending on what Fabio's 500 variables actually seem to measure, might a
reduction of that number using factor analysis be a possibility? If so, the
three above mentioned sources are still needed very much...!
* For searches and help try: