Announcement

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

  • Drawing observations from bivariate heteroskedastic normal

    Hi,

    as the title suggests I was wondering if there's any way in Stata to generate two variables that are drawn from a bivariate normal distribution in which the variance of each variable has its own form of heteroskedasticity. In the case that the variables are independent you can generate a variable for the heteroskedastic standard deviation, and use that variable within rnormal() to generate the heteroskedastic variable. For homoskedastic variables you can use drawnorm, but I can't seem to be able to find a way to do both. The purpose is to simulate Tobit Type II data, where both the variance in the selection and value processes are heteroskedastic.

    Thank you for any help or direction in this matter!
    Alfonso Sanchez-Penalver
Working...
X