Announcement

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

  • Panel data analysis for brand choice

    Cracker clogit.xlsx Hi, I got a question regarding panel data to run a conditional logistic regression. I have already run the conditional logistic regression but i want to confirm whether i did it correctly or not because i'm getting errors for sum commands which i really need to do for a complete analysis ( for instance defining time variable to create lags) it says that "repeated time values" in panel. Could you please help me with this because i need to include lagged variables to capture dynamic effects? I have done my research on Google but couldn't find the solution because I didn't see any data set related to the one i got. Therefore, i would be grateful if any one could help me through!
    I have a data on consumer choices for 10 purchase occasions; say i = 1......1360, t= 1.....10 and j= 1....4 (Individuals, choice occasions and alternatives). Since it's a panel data with a categorical dependent variable, therefore for conditional logistic I created a long form structure like (1 individual * 4 choices available) plus each individual was observed for 10 different occasions so (1*4*10) 40 rows related to each individual where four rows represent one choice occasion.
    Thank You!

    Note: (kindly, have a look on data set i attached or you can see a screen shot for explanation)
    Last edited by M A Ahmad; 05 Apr 2014, 04:22.

  • #2
    This is not the forum for answering statistical/programming questions.

    Use this forum to make test posts, practice with the formatting features of the forum software, etc.
    Delete this question and re-post in the "General Forum".
    Last edited by Steve Samuels; 05 Apr 2014, 16:48.
    Steve Samuels
    Statistical Consulting
    [email protected]

    Stata 14.2

    Comment


    • #3
      Sorry, I forgot that you can only delete up to an hour after you post your question. So, just re-ask in the general forum.

      Steve
      Steve Samuels
      Statistical Consulting
      [email protected]

      Stata 14.2

      Comment

      Working...
      X