st: Suest Command on an Unbalanced Panel Data to compare coefficients of two groups.

[natasha agarwal <agarwana2@googlemail.com>]

Dear Everyone,

I am estimating an augmented production function on an unbalanced
panel data set to test whether the impact of FDI differs across two
groups, namely developed and developing countries.

For similar reasons, I manually transform the data by using the
'center' command and then use the 'suest' command followed by the
'test' command to check for equality of coefficients on the FDI
variable across the two groups.

bysort number: center lnrval lnk lnw lnvfdi y14-y18, casewise
regress c_lnrval c_lnk c_lnw c_lnvfdi c_y14-c_y18 if dev1==1
est sto fe2
reg c_lnrval c_lnk c_lnw c_lnvfdi c_y14-c_y18 if dev1==0
est sto fe3
suest fe2 fe3, vce(cluster pronum)
est sto fe4
test [fe2_mean]c_lnvfdi==[fe3_mean]c_lnvfdi

dev1==1 if the country if developed and zero otherwise.

I estimated the same equation (but on an uncentered data) but this
time with the xtreg command.

xtreg lnrval lnk lnw lnvfdi y14-y18 if dev1==1, fe vce(cluster pronum)
xtreg lnrval lnk lnw lnvfdi y14-y18 if dev1==0, fe vce(cluster pronum)