Re: Logistic regression and linear regression
- From: Greg Heath <heath@xxxxxxxxxxxxxxxx>
- Date: Thu, 5 Jun 2008 02:22:05 -0700 (PDT)
On May 26, 9:44 am, maria.hartb...@xxxxxxxxx wrote:
Hello,
Would I get the same results if I turned my binary data into logratios
and then run a linear regression rather than running a logistic
regression?
The reason for this is that I need to run stepwiste forward logistic
regression in matlab but matlab has no function for that so I thought
that maybe I could turn my binary variables to log ratios and then do
a linear.
PS. Maybe the question seems simple but I am not a statistics person.
MATLAB now has a function that will do that. See
the documentation in the MATHWORKS website for
details.
If the new version is not accessible, it would be
beneficial to know what variables are chosen via a
linear fit.
However, beware of the blind use of STEPWISE or STEPWISEFIT.
Although you can use plain vanilla forward and backward
search, it is better to use your a priori knowledge of the
problem to specify both a starting set of variables and
a subset of those that must be kept.
Hope this helps.
Greg
.
- Follow-Ups:
- Re: Logistic regression and linear regression
- From: Greg Heath
- Re: Logistic regression and linear regression
- Prev by Date: Re: Bootstrapping & Structural Equation Modeling
- Next by Date: Re: Logistic regression and linear regression
- Previous by thread: Afonso´s conjecture formula
- Next by thread: Re: Logistic regression and linear regression
- Index(es):
Relevant Pages
|
Loading