Announcement

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

  • Simultaneous equation (reg3) and clustered standard errors

    Hi everyone,

    Does anyone know how to obtain clustered standard errors when using reg3 or sureg? I've looked online and there doesn't seem to be a straightforward solution. Bootstrapping alone does not work either-- the clustering is key.

    Thanks!

  • #2
    I am aware of how to cluster when using sureg; I came to this post looking for a solution to clustering when using reg3. So here's the solution to clustering when using sureg: use a slightly different command - suest - which allows for clustering. Say I have two regressions where I expect the errors to be correlated (which is why I am running sureg in the first place.) Run the first regression and then store the estimates, e.g. estimates store r1. Then run the second regression and store the estimates again, e.g. estimates store r2. Then run suest r1 r2, vce(cluster id). For example, when clustering standard errors by state, I run - suest r1 r2, vce(cluster state).

    Comment


    • #3
      Hi Everyone,

      For clustering with SUR, I would suggest the Stata ado file mysureg from William Gould, Jeffrey Pitblado, and Brian Poi, which can be downloaded here: http://www.stata-press.com/data/ml4.html. It allows for use with svy commands, clustering and weighting.

      By the way, has anyone found a way to cluster with 3SLS? I would like to use something like reg3 because I am estimating an mlogit with an IV and gsem is prohibitively slow due to many fixed effects. Without an IV, I can define log-odds ratios of my 4 mutually exclusive outcome variables and run mysureg of the 3 log-odds ratios on my x-variables as follows:

      svyset clustervar [pw=weight]
      svy: mysureg (logodds_eq1 $xvars ) (logodds_eq2 $xvars) (logodds_eq3 $xvars)

      As far as I understand, this is now clustering and weighting my estimation.

      I need to jointly estimate these equations with SUR so that I can access the estimation results for all three equations simultaneously in order to compute marginal effects. Now, if I add an IV to this setting, I am looking for a command that does for the 3SLS what the mysureg does for SUR. Any ideas?

      Thanks!
      Cortnie
      (Stata 14.2)

      Comment

      Working...
      X