Announcement

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

  • Heteroscedasticity and Autocorrelation in Panel Data

    Hello everyone,

    I have a panel data set and would like to investigate in my model the impact of corporate venture capital (measured as the number of CVC activities) on innovation performance (measured as the number of patents) of family firms. The literature shows that the negative binomial regression has often been used to relate corporate venture capital to innovation performance. So of course it makes sense to use this form as well.

    But I don't read anything in the journals about testing for autocorrelation or testing for heteroscedasticity.
    My question now is: How can I test if this is present and how would I deal with it?

    Here is my regression (, fe - by Hausmann test).

    Code:
    xtnbreg N_Totalpatents_w c.No_Comp_lag1_w##Family_Firm firm_size_loglag1 xrdat_wlag1 industry_q_w_lag1 os_w_lag1 i.fyear , fe
    margins Family_Firm, dydx(No_Comp_lag1_w)
    Thanks a lot for your help!
    Sven

  • #2
    Sven:
    the following Stata thread is probably of your interest: https://www.statalist.org/forums/for...roskedasticity
    Kind regards,
    Carlo
    (StataNow 18.5)

    Comment

    Working...
    X