Announcement

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

  • How can I Correctly Build the Fixed Effects Multinominal Logistic Regression Model with Panel Data?

    Hello,
    I ran into difficulties when building fixed effects Multinominal Logistic Regression Model with Panel Data using Stata.
    The dataset is a longitudinal and the variable "StudentID‘’ is nested within the variable "SchoolID".
    Meanwhile, each student was measured multiple times and the variable "time" is in the dataset.
    My boss wants me to do a fixed effect multinominal logistic regression model for him and he said he knew nothing about it. (His boss asked him to do this)

    The 1st question is when using xtset command, for this dataset, I should use xtset studentID or xtset studentID time or xtset schoolID?

    The 2nd question is when buidling the requested fixed effect model,
    what command I should use?
    xtmlogit depvar indvar1 indvar2 indvar3 indvar1#indvar2, fe vce(cluster schoolID) rrr is correct?
    when adding (cluster schoolID) into the model, the error message said, "groups are not nested within clusters".

    Thank you for your help.

  • #2
    Can someone help? Thank you.

    Comment


    • #3
      Can someone help me to write the correct code? Thank you.

      Comment


      • #4
        Can someone help me to write the correct code? Thank you.

        Comment


        • #5
          I think your boss just want to see the crude results using Multinominal Logistic Regression Model without considering mixed effects.
          xtmlogit is for mixed effects logistic model. you just do 'mlogit' first.

          Comment


          • #6
            Thank you!

            Comment


            • #7
              xtset schoolID
              xtmlogit depvar indvar1 indvar2 indvar3 indvar1#indvar2, fe

              Comment

              Working...
              X