Announcement

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

  • Split Panel Data Set into Subsamples

    Hi there,
    I have a panel data set comprising of 118 companies (range 10 years). the data set consist a corporate governance index with minimuim value of 0 to maximium value of 10. i want to split the data based on based on the corporated index into two sub-samples of low level corporate governance medium level corporate index and high level of corporate index.
    can anybody guide me how to split the panel data set into three subsamples based on the corporate governance index variable. thanks in advance

  • #2
    Waqar:
    assuming that you have no missing values, you may want to try something along the following lines:
    Code:
    gen flag=0 if gov_index<=3
    replace flag=1 if gov_index >3 & gov_index<=6
    replace flag=2 if gov_index>6
    Kind regards,
    Carlo
    (Stata 18.0 SE)

    Comment


    • #3
      What's wrong with just using the index as a quantitative predictor?

      Comment


      • #4
        Originally posted by Carlo Lazzaro View Post
        Waqar:
        assuming that you have no missing values, you may want to try something along the following lines:
        Code:
        gen flag=0 if gov_index<=3
        replace flag=1 if gov_index >3 & gov_index<=6
        replace flag=2 if gov_index>6
        Dear Carlo Lazzaro,
        As the sample is panel data so 10 years of data is collected regarding various variables related to each firm (total firms 118). Do i have to take avg of corporate governance (10 years' values) of each firm before applying the above code. as i think the above code will result in disintegrating the group of as year wise observation in respect of each firm and the resultant subsamples may be unbalanced.

        Comment


        • #5
          Originally posted by Nick Cox View Post
          What's wrong with just using the index as a quantitative predictor?
          Governance index is not main predictor in my model. Instead i want to evaluate the behavior of my main predictor in the firms with low, medium and high level of governance.

          Comment


          • #6
            Still the same question. Governance index can be regarded as a secondary predictor. Degrading it to 3 categories is throwing away information.

            Comment

            Working...
            X