| Date: | Wed, 16 Aug 2006 09:20:30 -0500 |
|---|
| Reply-To: | Gary Rosin <grosin@stcl.edu> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | Gary Rosin <grosin@stcl.edu> |
|---|
| Subject: | Re: Loss function for log-likelihood nonlinear regression of
proportion via Inverse Logit |
|---|
| In-Reply-To: | <002901c6c126$b1ffd7d0$a200a8c0@NOTEBOOK> |
|---|
| Content-Type: | text/plain; charset="us-ascii"; format=flowed |
|---|
At 06:25 AM 8/16/2006, Hector Maletta wrote:
> The expression for non linear expression for the pass rate mentioned in
>your message is exactly the definition of the probability of passing the
>exam in logistic regression. Calling Z to the combination
>b1X1+b2X2+...+bnXn, that probability is exp(z) / 1 + exp(Z), and the b
>coefficients are the log odds ratios coefficients computed by logistic
>regression.
Yes. In Nonlinear Regression that expression is what I use as the
'Model". The default loss function in least-squares. I'd like to use
log-likelihood instead, but don't know the expression for that loss
function.
Gary
---
Prof. Gary S. Rosin Internet: grosin@stcl.edu
South Texas College of Law
1303 San Jacinto Voice: (713) 646-1854
Houston, TX 77002-7000 Fax: (713) 646-1766
| 
