Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Multinomial logit with a nonlinear combination utility equation


From   "Simons, Chris N" <Chris.Simons@atkinsglobal.com>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: Multinomial logit with a nonlinear combination utility equation
Date   Tue, 1 Nov 2011 13:15:57 -0400

I'm new to Stata and I want to estimate a multinomial logit model where the logit utility equation is not a simple linear combination. Some of the variables to be estimated must be within a log operation. For example, the utility equation typically looks like this:

Util = p1*a + p2*b + ln(c + p3*d)

where p1, p2 and p3 are the parameters to be estimated and a, b, c and d are independent variables.

It's my understanding that standard solution algorithms may have a difficult time solving such a problem and that specialized algorithms may be necessary. I've looked at the -ml- command and trying to figure out how to program the problem including specify the log-likelihood functions, gradients and hessian's is daunting enough (for me - maybe not for many of you!). I'm concerned that even if I did manage to specify them correctly the solution algorithm may still struggle.

Is there an existing package (or other command) to handle such estimations? I have been unable to find any in my searches. Any advice would be appreciated.

I'm currently using Stata/SE 11.2 for Windows (32-bit)  Born 30 Mar 2011.

Thanks,
Chris

Chris N. Simons
Senior Transportation Planner

ATKINS

1600 RiverEdge Parkway, NW, Suite 600, Atlanta, GA, 30328 | Tel: +1 (770) 933-0280 | Direct: +1 (678) 247-2457 |
Cell: +1 (770) 653-3993 | Email: chris.simons@atkinsglobal.com | Web www.atkinsglobal.com/northamerica  www.atkinsglobal.com


This electronic mail communication may contain privileged, confidential, and/or proprietary information which is the property of The Atkins North America Corporation, WS Atkins plc or one of its affiliates. If you are not the intended recipient or an authorized agent of the intended recipient please delete this communication and notify the sender that you have received it in error. A list of wholly owned Atkins Group companies can be found at http://www.atkinsglobal.com/site-services/disclaimer

Consider the environment. Please don't print this email unless you really need to.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index