Hi everyone,
I am trying to run a panel data regression with time and firm fixed effects, but I am encountering some serious difficulties which i can't seem to solve (up until now). I hope you have the solution for me. I am studying the causal effect of Return On Assets and Return On Equity on Executive Compensation for US Commercial and Investment Banks. When I try to define the FE regression I do: xtset BANKTYPE FiscalYear and I get the following message:
repeated time values within panel
r(451);
My question is : How can I run the below regression with time FE (taking the variable FiscalYear (2000,...2018) as time effect) and firm FE (taking the variable BANKTYPE as "industry" as this highlights Commercial banking or Investment banking) properly? The "Group variable" should be BANKTYPE and the "numbers of groups" should be 2 (Commercial and Investment banks). Please refer below for details of my DataSet and regression.
These are the details of my Panel Dataset:
Period: 2000-2018
#Datapoints: 11,343
# Banks: 171
Regression: COMPij,t = B0 + B1*PERFi,t + B2*AGEij + B3*Log(TA)i,t + B4*Log(EMPL)i,t + B5*MTBRi,t + B6*LEVi,t + B7*BANKTYPE + B8*FiscalYear + uij,t
Where:
COMPij,t = Executive Compensation of Executive j at company i at the end of FiscalYear t , t = 2000,....,2018
PERFi,t = Financial Performance of company i at the end of FiscalYear t, (for example, ROA or ROE)
AGEi,j = Age of Executive j at company i
Log(TA)i,t = natural logarithm of Total Assets of company i at the end of FiscalYear t
Log(EMPL)i,t = natural logarithm of the number of Employees of company i at the end of FiscalYear t
MTBRi,t = Market-to-Book Ratio of company i at the end of fiscal year t
LEVi,t = Leverage company i at the end of fiscal year t
BANKTYPE = 1 if US Commercial bank , 0 if US Investment bank (Dummy variable)
FiscalYear = 2000,.....,2018
uij,t = error term
When I try to define the FE regression I do: xtset BankType FiscalYear and I get the following message:
repeated time values within panel
r(451);
0 Response to Panel Data regression with time and firm fixed effects
Post a Comment