Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • test of overidentifying restrictions after fracivp


    Dear all,

    I am currently analyzing a model in which the dependent variable is a proportion.
    In particular. in order to account for endogeneity, I would like to use the following command taken from https://www3.nd.edu/~rwilliam/stata

    code:
    fracivp

    After the estimates, I need to compute the test of overidentifying restrictions.
    Do you know if there is a routine or a code to do it?

    For example, after using ivtobit, the test of overidentifying restrictions can be computed by:

    code
    overid


    Many thanks in advance

  • #2
    First off, the correct link is

    https://www3.nd.edu/~rwilliam/stats3...onseModels.pdf

    fracivp is a hacked version of ivprobit. It relaxes the requirement that the dv be binary and lets it instead range from 0 to 1. So, if there is a way to do it with ivprobit you may be able to do it with fracivp.

    Also, as the handout also points out, cmp can estimate the same models. So, if there is a way to do it with cmp, you can use that.

    In short, only a few people in the world know that fracivp even exists since I have Never officially released it. But if you check out ivprobit or cmp you might find something there.
    -------------------------------------------
    Richard Williams, Notre Dame Dept of Sociology
    Stata Version: 15.1MP (2 processor)

    EMAIL: rwilliam@ND.Edu
    WWW: https://www3.nd.edu/~rwilliam

    Comment


    • #3
      Ok, I will try to work with cmp. Thanks for your reply.

      Comment

      Working...
      X