Fitting a logistic regression using lassoglm in matlab

Asked Mar 22 '13 at 15:05

Active Feb 06 '14 at 21:03

Viewed 186 times

I am fitting a logistic regression model using lassoglm in matlab. I issued the following command

[B,FitInfo] = lassoglm(ec50,Y,'binomial','Lambda',0.0001,'CV',10);

Now once I get the coefficients B, how am I supposed to use it. I mean do I have to do

$P(y=1) = \frac{e^{(\theta^tx)}}{1 + e^{(\theta^tx)}}$ where $\theta$s are obtained from B. Is it the case?

asked Mar 22 '13 at 15:05

user34790

Other threads bearing on this problem can be found by searching our site. – whuber Feb 06 '14 at 21:12
B is essentially the weights. You can use ec50*B to get estimates for the fit of Y. – mugs Feb 06 '14 at 21:03

0 Answers0