Announcement

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

  • Year and Industry Fixed Effects in Cross-sectional logit regression

    Hi Everyone,
    I would appreciate your help on the below:
    I am using Stata 17. I have a cross-sectional data of companies that made an acquisition and their corresponding target company that got acquired.
    I have company financials as independent variables, most are numbers, some are dummy variables (eg. Hostile takeover or not. Majority Acquisition or not, etc.).
    My dependent variable(s) is the acquisition payment method (binary 1 or 0). I am running the logistic regression 3 separate times for the 3 payment methods (Cash payment, Stock payment or Mixed payment).
    I want to include industry and year fixed effects.
    The command I'm using is: logistic Y X1 X2 X3 X4 i.IndustryCode i.Year, vce(cluster AcquirorCode)
    For the Mixed Payment dependent variable, I have no issues in adding both fixed effects.
    For the Cash Payment dependent variable, I have no issues with running the regression either with i.IndustryCode or with i.Year, but when I add both, then i am getting an error: convergence not achieved r(430).
    For the Share Payment dependent variable, I have no issues with i.IndustryCode. But when i run a regression with just i.Year or with both fixed effects, I get the same error.
    For the Share Payment, I tried running clogit Y X1 X2 X3 X4, group(Year) and I got the same error. Also, with clogit I don't think I can add 2 groups (Industry and Year), and cluster by AcquirerCode correct?
    I am sure the variables IndustryCode and Year are integers.
    Any idea why this is happening?
    Regards,
    Paul K.

  • #2
    Paul:
    welcome to this forum.
    Your post is expected to receive no replies as your (too) long description does not offer any detail for any helpful hint.
    Please, spend some minutes of your time to read (and subsequently act on) the FAQ. Thanks.
    Eventually, if your query implies class/homewoks, I'm afraid this is not the right webiste to post your questions (please see https://www.statalist.org/forums/help#adviceextras, #4).
    Kind regards,
    Carlo
    (Stata 19.0)

    Comment


    • #3
      Thanks Carlo,
      This was for my PhD dissertation. But i believe i figured it out.
      Regards,
      Paul K.

      Comment

      Working...
      X