Panel data with repeated time values within panel r(451); due to multiple directorships

[Emil Holm]

Dear all,

I am writing my Master's thesis in Economics, and I am working on a data set on board of directors in the US over the past 10 years. Each director has a unique ID, and I can follow the director year-by-year. The problem arises, as I get the "repeated time values within panel r(451);" error when I try and set up my panel data.

This, I believe, is because a director can hold multiple directorships in different companies at the same time, why the year and the director ID is not unique as a match. To give an example, a director with ID "503" can be a director on the board of both "Eastman" and "TE" in year 2011.


data.tiff

Is there a smart workaround? At the moment, I have the following code:

global t Time
global id ID
global ylist PD
global xlist Women ROA MTB LIQUID SOLV RISK IndD

*isid ID Time (to check for uniqueness)

describe $id $t $ylist $xlist
summarize $id $t $ylist $xlist

* Set data as panel data
sort $id $t
xtset $id $t
xtdescribe
xtsum $id $t $ylist $xlist

* Pooled OLS estimator
reg $ylist $xlist

* Population-averaged estimator
xtreg $ylist $xlist, pa

* Between estimator
xtreg $ylist $xlist, be

* Fixed effects or within estimator
xtreg $ylist $xlist, fe


Thanks in advance,

Emil

[Carlo Lazzaro]

Emil:
1) if you do not plan to use time series related commands, such as lags and leads, you can simply -xtset- your dataset with your -panelid- only:

xtset panelid

2) trying so many regression model specifications is unfruitful: skim through the literature in your research fielad and see what others did in the past when presented with the same research goal.
Moreover, the between estimator is seldom used (the random effect specification is frequent, instead). Please also note that there tests that can help you comapring -fe- vs -re- specification (see -help hausman-)..

[Emil Holm]

Thanks, Carlo.

I tried -xtset- only with the panelid and I got a new error "must specify timevar; use xtset r(459);"
Do you know what went wrong?

And I intend on performing a Hausman test later on, but you are right, I should limit myself to re and fe.

Sincerely,
Emil

[Carlo Lazzaro]

Emil:
could you plase show what you typed and what Stata gave you back? Thanks.