Questions about Problem Sets

[alex-hoagland]

If you have a question about a problem set, please add it here!

[YICHI525]

Hi professor Alex,
I saw a lot of abbreviations in the assignment dataset. I am so sorry that I think I could not understand the meaning of these abbreviations correctly. Is the LEBF means the law Enforcement Benevolent Fund? What about the LEBM and INFM and the others? I have googled them, but I did not find something useful.

[alex-hoagland]

Hi Yichi, the relevant abbreviations are defined in the problem set itself -- I'd encourage you to look there. Feel free to ignore additional covariates with undefined abbreviations -- if you're curious I'd be happy to help identify the variables, but they aren't necessary for this problem set.

[YICHI525]

Professor Alex, thank you for your reply.
I think the first step to draw a DAG is that I know the meaning of the variates? As you have shown in the slides--Y means outcome and D means treatment and X means covariate, thus we could find their relationship. I think we should know the meaning of these variates and put them into different location and find their relationship. Is it correct? Because I found that if I do not know the meaning, I could not recognize which one is the outcome.

[alex-hoagland]

As you'll see in the assignment text, you don't have to use all the covariates, just the ones that seem relevant to you, so you can skip over any covariates that you don't understand. The assignment text lists which variable is the outcome (LEBF) and the main covariates of interest (GDPPC, HXPC, fertility)

[YICHI525]

Professor Alex, thank You for your reply.
Because English is not my native language, I found that I did not understand the problem in the assignment very well. I am so sorry to have disturbed you for asking these questions. I thought that "any other covariates" --this expression means that I need to use all the covariates. And I thought you just give us the variate names and let us to recognize which one is the outcome. It’s all my fault. I misunderstood the question. I am so sorry.

[alex-hoagland]

No problem at all! That’s what this discussion board is for. Glad that we were able to clarify it.

…

---------------------------------------------------------------------------- Alex Hoagland Assistant Professor of Health Economics Institute of Health Policy, Management, and Evaluation Dalla Lana School of Public Health University of Toronto https://alex-hoagland.github.io From: YICHI ***@***.***> Sent: Thursday, September 8, 2022 9:28 AM To: ***@***.***> Cc: Hoagland, ***@***.***>; ***@***.***> Subject: Re: [alex-hoagland/HAD5744_2022F] Questions about Problem Sets (Discussion #4) Professor Alex, thank You for your reply. Because English is not my native language, I found that I did not understand the problem in the assignment very well. I am so sorry to have disturbed you for asking these questions. I thought that "any other covariates" --this expression means that I need to use all the covariates. And I thought you just give us the variate names and let us to recognize which one is the outcome. It’s all my fault. I misunderstood the question. I am so sorry. — Reply to this email directly, view it on GitHub<#4 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AK62K2RAHWYAF2WAAWZH7OLV5HSYLANCNFSM542YBQEA>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[CarterJasmine]

Dear Professor Alexander Hoagland,

If the Matchit package by Ho and colleagues 2011 is not available in Stata - what is recommended instead?

I have used the manual functions treatment effects --> continuous outcome --> propensity-score matching as well as .pscore then .psmatch2.

Also, do you recommend including the standardized mean difference in the balance tables?

Thank you very much.
I hope that you have a lovely evening!
Kind regards,
Jasmine Carter

[alex-hoagland]

Hi Jasmine,

In Stata, I usually use the "teffects" command with the psmatch subcommand (for propensity score matching). It should give you a similar outcome to what you have done but may be easier to implement.

The standardized mean would be a great idea to include in the balance tables.

Let me know if you have other q's!

[CarterJasmine]

Dear Professor Alexander Hoagland,

Thank you very much for the clarification!

Kind regards,
Jasmine

[alex-hoagland]

Hi Jasmine, In Stata, I usually use the "teffects" command with the psmatch subcommand (for propensity score matching). It should give you a similar outcome to what you have done but may be easier to implement. The standardized mean would be a great idea to include in the balance tables. Let me know if you have other q's! ————————————————————— Alex Hoagland, PhD Assistant Professor of Health Economics Institute of Health Policy, Management, and Evaluation Dalla Lana School of Public Health University of Toronto https://Alex-Hoagland.github.io

…

________________________________ From: CarterJasmine ***@***.***> Sent: Tuesday, November 8, 2022 10:22:30 PM To: alex-hoagland/HAD5744_2022F ***@***.***> Cc: Hoagland, Alexander ***@***.***>; Author ***@***.***> Subject: Re: [alex-hoagland/HAD5744_2022F] Questions about Problem Sets (Discussion #4) Dear Professor Alexander Hoagland, If the Matchit package by Ho and colleagues 2011 is not available in Stata - what is recommended instead? I have used the manual functions treatment effects --> continuous outcome --> propensity-score matching as well as .pscore then .psmatch2. Also, do you recommend including the standardized mean difference in the balance tables? Thank you very much. I hope that you have a lovely evening! Kind regards, Jasmine Carter — Reply to this email directly, view it on GitHub<#4 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AK62K2WQ5ZSAQQLIPQBOTBTWHMKHNANCNFSM542YBQEA>. You are receiving this because you authored the thread.Message ID: ***@***.***>