For a panel data analysis with 6 million observations, I will receive the following warnings but when I check for multicollinearity, it's ok.
Do you have any comments on how to deal with these warnings?
Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
warning: missing F statistic; dropped variables due to collinearity or too few clusters
Warning: Variance matrix is nonsymmetric or highly singular.
If it helps, here is the model:
reghdfe LogHourlyReadingTime USDRate InternetShutdownFlag LaggedCumulativeBooks, absorb(DayOfWeek WeekNumber HourOfDay UserID) cluster(DayOfWeek WeekNumber HourOfDay UserID)
When I run it on my Stata (SE18 and BE17), the results table is empty and I only see coefficients. But on some other devices, it generates the result table but without F-statistics.
I need your help a lot!
Thanks!
Do you have any comments on how to deal with these warnings?
Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
warning: missing F statistic; dropped variables due to collinearity or too few clusters
Warning: Variance matrix is nonsymmetric or highly singular.
If it helps, here is the model:
reghdfe LogHourlyReadingTime USDRate InternetShutdownFlag LaggedCumulativeBooks, absorb(DayOfWeek WeekNumber HourOfDay UserID) cluster(DayOfWeek WeekNumber HourOfDay UserID)
When I run it on my Stata (SE18 and BE17), the results table is empty and I only see coefficients. But on some other devices, it generates the result table but without F-statistics.
I need your help a lot!
Thanks!
