Can I use Stata as Mail Merge? Is there native functionality or a user written package for this? Can a user written package be created?

[Joro Kolev]

Good afternoon,

If I have a dta file like this:

* Example generated by -dataex-. To install: ssc install dataex
clear
input str11 name str19 email float(midtermgrade finalgrade)
"John Smith"  "[email protected]" 78 98
"Betty White" "[email protected]" 35 29
end

Can I use Stata to send separate emails to Betty White and John Smith containing their, and only their respective grades on the midterm and final exams?

[Leonardo Guizzetti]

i've never used Stata for a mail merge, but in principle a solution can be worked out. To my knowledge, Stata does not natively support email. However, if you're on a Windows machine you can write a Stata program that output Powershell scripts (.ps1 files) that contain the contents and header info of your email, and then execute those by triggering a shell command. I can dig up my simple powershell script to show the basics if you would like.

[Leonardo Guizzetti]

It seems that there are many possible solutions by programs already on the SSC (-search send email-), most of which revolve around powershell. See -email-, -emailme-, -sendmail-, -psemail-, -mail-.

[Joro Kolev]

Thank you very much, Leonardo !

I looked through the packages you listed, and only -psemail- allegedly does what I have on my mind.

I will experiment with this one and see whether it really works or not.

It seems that there are many possible solutions by programs already on the SSC (-search send email-), most of which revolve around powershell. See -email-, -emailme-, -sendmail-, -psemail-, -mail-.

[Steven Dee]

Joro,
Your posts have been very informative. Allow me to ask some basic questions.
1. Is it meaningful to consider cointegration in a logit?
2. The predictor variables are non-stationary in levels but stationary in differences. Since they are all I(1), is it meaningful to test for cointegration between them rather than differencing them?

--
Steven Dee

[Leonardo Guizzetti]

Joro,
Your posts have been very informative. Allow me to ask some basic questions.
1. Is it meaningful to consider cointegration in a logit?
2. The predictor variables are non-stationary in levels but stationary in differences. Since they are all I(1), is it meaningful to test for cointegration between them rather than differencing them?

--
Steven Dee

Steven, it's considered bad form to resurrect an old thread with new and unrelated questions. Please create a new thread with your questions. Be sure to take time to read the FAQ, especially section 12 with how to ask questions. While your questions don't seem related to asking for help with code, you may want to include more context about your problem. Lastly, it's generally a good idea to address the question to anyone on the forum, not someone in particular, as any one person may not respond or have gone away from the forums.