Hello,
I want to identify company characteristics that have an impact on the size of dividends firms pay. In my regression I want to use time- and industry-fixed effects (do you think that it would make sense to include firm-fixed effects instead of industry-fixed effects? Fixed-effects confuse me a little bit, therefore I'm unsure whether my approach makes sense). To test whether using fixed effects makes sense I want to perform the Hausman test. Therefore, I first must declare my data set as panel data.
The command "xtset sic_code year" does not work, I get the error message "repeated time values within panel". So I just wrote: xtset sic_code.
Then I used xtreg and plugged in my dependent variable "dvt" (for Dividends) and my independet variables like profitability age etc. and added i.year to account for time-fixed effects and then specifying the fe option: xtreg dvt prof age i.year, fe
Then I stored the results (estimates store fe) and did the same regression just with the re option and again storing the results.
When I then perform the Hausman test (hausman fe re) I get a warning: "the rank of the differenced variance matrix (22) does not equal the number of coefficients
being tested (23); be sure this is what you expect, or there may be problems computing the test. Examine the output of your estimators for anything unexpected and possibly consider
scaling your variables so that the coefficients are on a similar scale."
chi2 < 0 ==> model fitted on these data
fails to meet the asymptotic assumptions
of the Hausman test; see suest for a
generalized test.
I dont know what I did wrong and how to fix this error. I would appreciate every help!
I want to identify company characteristics that have an impact on the size of dividends firms pay. In my regression I want to use time- and industry-fixed effects (do you think that it would make sense to include firm-fixed effects instead of industry-fixed effects? Fixed-effects confuse me a little bit, therefore I'm unsure whether my approach makes sense). To test whether using fixed effects makes sense I want to perform the Hausman test. Therefore, I first must declare my data set as panel data.
The command "xtset sic_code year" does not work, I get the error message "repeated time values within panel". So I just wrote: xtset sic_code.
Then I used xtreg and plugged in my dependent variable "dvt" (for Dividends) and my independet variables like profitability age etc. and added i.year to account for time-fixed effects and then specifying the fe option: xtreg dvt prof age i.year, fe
Then I stored the results (estimates store fe) and did the same regression just with the re option and again storing the results.
When I then perform the Hausman test (hausman fe re) I get a warning: "the rank of the differenced variance matrix (22) does not equal the number of coefficients
being tested (23); be sure this is what you expect, or there may be problems computing the test. Examine the output of your estimators for anything unexpected and possibly consider
scaling your variables so that the coefficients are on a similar scale."
chi2 < 0 ==> model fitted on these data
fails to meet the asymptotic assumptions
of the Hausman test; see suest for a
generalized test.
I dont know what I did wrong and how to fix this error. I would appreciate every help!
Comment