Announcement

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

  • Hierarchical Linear and Logisitic Regression Models with Random Coeffcients

    I am currently attempting to run HLM models that will produce random coefficients for both level-1 (offender and case characteristics) and level-2 (state contextual factors) independent variables. The commands I have used thus far are the following: xtmixed (or xtemlogit) DV IVs || FIPS_Code:, variance mle. I know that running these commands only gives me a random intercept. I tried to run these models where I include all the independent variables in the random part, but Stata does not produce anything.

    Is there a different command for running HLM models where the coeffiicients of all the independent variables are entered into the model as random?

    Additionally, is it possible to produce standardized beta coefficients for multilevel models, such as HLM?

    Thank you for your assistance.

  • #2
    Welcome to Statalist. I would suggest you look at the FAQ on asking questions. - providing Stata code in code delimiters, readable Stata output, and sample data using dataex will help us help you. Also, don't use exclamation points and bold - they just annoy folks who might help you. People on this list are helping out of the goodness of their hearts - many of them don't like to be pushed (and bold and exclamation points feels pushy).

    It is not clear what you mean by having all the variables with random coefficients. This certainly can be done but you have to have some constraints on the parameters. You should (i) google Stata random coefficients, (ii) enter findit random coefficients in the command line of Stata, and (iii) look in the subject index for random coefficients. I don't do random coefficient models, but you might look at xtrc .

    Comment


    • #3
      Thank you for your feedback.

      Comment

      Working...
      X