Announcement

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

  • "matrix for _cmp_y2 equation appears ill-conditioned" message using cmp

    Dear Stata users,

    I am using the cmp command for the first time. I am trying to estimate a two equation model like Y = a1X1 + ...... and X1 = b1Z1 + .... [with x1 needing an ordered probit and Y a simple ols]

    When I run the estimation, I have the following error :

    regressor matrix for _cmp_y2 equation appears ill-conditioned. (Condition number = 1798.0022.
    > )
    This might prevent convergence. If it does, and if you have not done so already, you may need to remov
    > e nearly
    collinear regressors to achieve convergence. Or you may need to add a nrtolerance(#) or nonrtolerance
    > option to the command line.
    See cmp tips.



    I've read in other posts that this is caused by the size of xx'. Though, I was not able to find out the exact consequences of this kind of warning.

    In my situation, the estimation displays 5 iterations. Does it mean that the process has converged so I can neglect the warning message ? Or may it mean that the process has not converge thus potentially leading to wrong estimates ? Since I am not familiar with the command, I would prefer to stick to de default option with respect to nonrtolerance.

    Iteration 0 exhibits a "(non concave") message. I do not know if it has to do with it.

    Many thanks in advance

    Eric

  • #2
    The message states the possible consequence: failure to converge. In your case, that is not happening.

    Comment


    • #3
      Dear David.
      Many thanks for your quick answer. Then, I guess that if no converge occurs, a specific message is displayed.
      Thanks again,
      Eric

      Comment

      Working...
      X