> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of
> Stephen Choularton
> Sent: Tuesday, June 07, 2005 11:49 PM
> To: R Help
> Subject: [R] logistic regression (glm binary)
>
>
> Hi
>
> I am looking for a couple of pointers using glm (family = binary).
Do you mean "binomial" instead of "binary"?
> 1. I want to add all the products of my predictive features as
> additional features (and I have 23 of them). Is there some easy way to
> add them?
Probably something along the line:
> dat <- data.frame(y=sample(0:1, 100, replace=TRUE), matrix(runif(300),
ncol=3))> fm <- glm(y ~ .^2, family="binomial", data=dat)
> summary(fm)
Call:
glm(formula = y ~ .^2, family = "binomial", data = dat)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.654 -1.175 0.608 1.116 1.651
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 3.264 1.536 2.125 0.0336 *
X1 -3.379 2.026 -1.668 0.0953 .
X2 -4.659 2.244 -2.077 0.0378 *
X3 -3.531 2.060 -1.714 0.0865 .
X1:X2 4.535 2.775 1.634 0.1022
X1:X3 2.123 2.639 0.804 0.4212
X2:X3 4.315 2.746 1.571 0.1161
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05
'.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 137.99 on 99 degrees of freedom
Residual deviance: 131.84 on 93 degrees of freedom
AIC: 145.84
Number of Fisher Scoring iterations: 3
> 2. I want to drop each feature in turn and get the most significant,
> then drop two and get the next most significant, etc. Is there some
> function that allows me to do this?
Not that I know of, and most likely for a very, very good reason...
Andy
> Thanks
>
> Stephen
>
> --
> No virus found in this outgoing message.
> Checked by AVG Anti-Virus.
>
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
>
>
>