Comparing fixed effects coefficients for several dependent variable

[Narumi Iihara]

Dear all, I have estimated a fixed effect panel regression model for several dependent variable in my data set.

For example, I have:
xtreg y1 x1 x2 x3, fe cluster (ID)
xtreg y2 x1 x2 x3, fe cluster (ID)
xtreg y3 x1 x2 x3, fe cluster (ID)
xtreg y4 x1 x2 x3, fe cluster (ID)
xtreg y5 x1 x2 x3, fe cluster (ID)
xtreg y6 x1 x2 x3, fe cluster (ID)

and I would like to know if x1 (coefficient) in all regression is statistically significant or not,
so I try to use "suest" command and do multiple hypothesis testing.

For example,
test (x1), mtest
test (x1), mtest (bonferroni)
test (x1), mtest (holm)

However stata gives me the comment "xtreg is not supported by suest".
So I couldn' t do the multiple hypothesis testing.
I have looked everywhere for a test on stata but I only got more confused by the various different answers.
Can anyone help me and provide me with an easy fix for this problem?

Thank you very much.
Narumi

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions.
It is not clear what such a test would mean - normally the different y's will have different standard deviations and scales so saying the parameter on x in influencing (say) sales differs from the parameter on x in influencing (say) leverage is not very meaningful. If you restated sales to be in $1000 dollars, you'd get different results.
If you really want to do this, you can do regress with i.ID as a regressor and clustered standard errors which is the same as xtreg. You can use sureg.