outreg2 for ivreghdfe

[Ken Yang]

Dear Statalists,

I am quite new to reghdfe and ivreghdfe. I hope someone can help me with my code.
I have a panel data with 800 firms of 40 industries in 10 years.

I want to run 2sls, z directly affects x, but indirectly affects y.

xtset firm year

ivreghdfe y (x = z) $controls, absorb(industry year) first
outreg2 using myfile.xls, replace stats(coef pval)

But I could not obtain both the 1st stage and 2nd stage results with outreg2. I have tried to search here, but there is no suggestion to use outreg2 for ivreghdfe.
Can anyone help me with this, please? I really appreciate your help.

Regards,
Ken

[Phil Bromiley]

You may need to estimate the first stage separately (visually check that what you're running is right, i.e., gives the same results as reghdfe), then work with that.

[Paul Singh]

Dear Statalists,

I am quite new to reghdfe and ivreghdfe. I hope someone can help me with my code.
I have a panel data with 800 firms of 40 industries in 10 years.

I want to run 2sls, z directly affects x, but indirectly affects y.

xtset firm year

ivreghdfe y (x = z) $controls, absorb(industry year) first
outreg2 using myfile.xls, replace stats(coef pval)

But I could not obtain both the 1st stage and 2nd stage results with outreg2. I have tried to search here, but there is no suggestion to use outreg2 for ivreghdfe.
Can anyone help me with this, please? I really appreciate your help.

Regards,
Ken

If you use/install estout you can do the following

 
 ivreghdfe y (x = z) $controls, absorb(industry year) first savefirst savefprefix(st1) esttab est1 st1*

However, now that I think about it. You can probably add savefirst and savefprefix(st1) to your command and then find out how it works with outreg. I've never really used outreg.

[iris lee]

Hello,

Although it might be too late but it might be useful to others. You could try the following command:

eststo:ivreghdfe y controls (x=iv)  , a(city year )  cluster(city) first savefirst savefp(f) // savefprefix(prefix) option stes a prefix for first stage estimation result
est restore fx // there must be "f"+"x" 
outreg2 using iv.doc,cttop(first) replace
ivreghdfe y controls (x=iv), a(city year ) cluster(city)
outreg2 using iv.doc,cttop(second)