Installing ivreg2 command without internet access

[Mara Hammerle]

Hello,

I have been using confidential data in a government remote desktop that doesn't have access to the internet. Hence, the easy option of installing new commands using "ssc install [command]" is not available. On the remote desktop, there are several ado packages that one can access. I have been trying to get the ivreg2 command to work as I need it for a journal article submission (it is the last step before the article will be accepted!). However I have had zero luck.

I have tried:
1) Run the ranktest.ado file followed by the ivreg2.ado file.
2) Set personal directory to the ivreg2 folder (containing the ado file and a sthlp file) and run the ranktest.ado file separately. (The ranktest folder containing the ranktest.ado file doesn't have any other files in it.)

Is there anything else I can try?

Thanks.

[Mara Hammerle]

I have now set up a folder containing all the relevant ado files (basically copy and paste-ing from the government-provided ado files in the remote desktop's library). I set the system directory "PERSONAL" path to the folder I created, using:

sysdir set PERSONAL "[name of folder containing ado files]"

I then used "sysdir" to check that it was connected properly, followed by:

sysdir which ranktest
sysdir which ivreg2

I then re-ran the command ivreg2. The error message was that I had not installed a recent-enough version of ranktest. However, the version required was older than the version Stata told me I had when I ran the "sysdir which" command.

Thus, I'm still very confused about how to proceed. Any help would be fantastic!

[Abdulwahab Bello]

Dear sir,
I am trying to estimate a system-GMM using the xtdpdgmm command. My dependent variable is co2 while my independent variables are; ict_ind re eu gdp gdp2 (gdp square) edu trad urb. I specify the model as; xtdpdgmm L(0/1). co2 ict_ind re eu gdp gdp2 edu trad urb, collapse gmm(co2, lag(2 4) model(diff)) gmm( ict_ind re eu gdp gdp2 edu trad urb, lag(1 3) model(diff)) ///
gmm(co2, lag(1 1) diff) gmm(ict_ind re eu gdp gdp2 edu trad urb, lag(0 0) diff) two vce(r)

Please is the above specification correct?

[Mara Hammerle]

Hi Abdulwahab,
I'm not sure why you're posting on my post as your question is completely unrelated.

[Davide Rigo]

Hi Mara,
I'm having the same issue. Have you had any luck?
Hope the paper is published by now!
Many thanks in advance.