[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Discriminant Analysis
You might want to consider using logistic regression as an alternative to discriminant analysis. Discriminant analysis assumes that all the predictors have a normal distribution, which is often not the case. If it is the case, discriminant analysis provides slightly higher statistical power, but since the assumption is often badly violated in one's data, this is often irrelevant. Logistic regression doesn't require making assumptions about the distributions of the predictors. David Greenberg, Sociology Department, New York University.
----- Original Message -----
From: Lei Wang <email@example.com>
Date: Thursday, October 17, 2002 5:50 am
Subject: st: Discriminant Analysis
> Dear Sir/Madam,
> I'm a student doing Economics in UK. Recent days I'm handling a
> regarding discriminant analysis, which needs STATA to deal with
> the data
> that I have.
> My project is about bankruptcy prediction. I have about 80
> financial statements, from which I'll use several ratios. They
> are: credit
> grade, % change in net cash cycle, gearing, net profit margin,
> ratio, liquid ratio. The 'credit grade' is from 1 to 14. What I'm
> going to
> do is to use discriminant analysis to distinguish those companies
> into two
> groups (above credit grade 11 and below credit grade 11) according
> to above
> As I'm just about to start to use STATA and I can't find enough
> information about Discriminant Analysis from STATA itself, could
> you please
> give me some detail information or advice about how to proceed it?
> It's an important and urgent case for me and your early response
> will be
> highly appreciated.
> Thank you very much.
> Lei Wang
> University of Bristol, UK
> Protect your PC - get McAfee.com VirusScan Online
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
* For searches and help try: