There are two available packages for estimating multinomial logit models in R, namely mnlogit and mlogit. I am wondering whether someone who used both can share his experience. Which one is better / more commonly used? Are there some advantages to using one of them? How do the packages deal with pannel data?

  • $\begingroup$ Discussion that require opinions belong at CV Meta. $\endgroup$ – Michael R. Chernick Apr 2 '17 at 12:19
  • 2
    $\begingroup$ @Michael Meta is only for discussing this site, not for any opinion-oriented discussions. Our chat is a good place for general discussions. In the present case, I can imagine that answers might refer to different algorithms or assumptions made by these packages and thereby could give this thread some useful statistical meaning, so I am reluctant to close it because it superficially deals with software packages. $\endgroup$ – whuber Apr 2 '17 at 22:12
  • $\begingroup$ searching on this site now, mnlogit has 7 hits, mlogit has hundreds ... $\endgroup$ – kjetil b halvorsen Mar 29 at 3:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.