[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: R: counterfactual model based on propensity score

From   "Carlo Lazzaro" <>
To   <>
Subject   st: R: counterfactual model based on propensity score
Date   Tue, 23 Jun 2009 07:59:49 +0200

Dear Statalisters,
I forward to you all Ilya Maria's posting in order to increase her
probabilities to receive a helpful hand.

Kind Regards,
-----Messaggio originale-----
Da: Ilya María Espino Cruz [] 
Inviato: martedì 23 giugno 2009 4.02
A: Carlo Lazzaro
Oggetto: counterfactual model based on propensity score

>Dear Carlo:

I got the Barbara Sianesi´s paper  that you recomended me, but the
mayority of commands of paper I can not use them. I have a Stata Version
9.2 and so when I use the "search" and "help" simply I couldn´t find them.

Some commands: psmatch, _times, _matchdif, _matchedid.

Can you help me to find them?



 Dear Maria,
> besides Martin's very useful reference, there's another .pdf on this topic
> (sorry, but I haven't the exact reference on hand)
> Barbara Sianesi. Implementing Propensity Score Matching Estimators with
> Prepared for UK Stata Users Group, VII Meeting, London, May 2001
> Kind Regards,
> Carlo
> -----Messaggio originale-----
> Da:
> [] Per conto di Ilya
> [iso-8859-1]
> María Espino Cruz
> Inviato: venerdì 19 giugno 2009 7.06
> A:
> Oggetto: st: counterfactual model
> Dear  Stata users:
> I am  researching about  targeting performance of program social. I have a
>  Household survey-income to make the calculations, I need to evaluate the
> performance of program social under counterfactual model based of perfect
> targeting.
> At moment I have made the followings steps:
> First: A probit model was estimated with all variables predicting whether
> a household receives the transfer of the program social.
> Second: The predicted probalities of receiving the transfer from the
> probit model were calculated. These predicted probabilities are the
> propensity scores.
> Now I need to make  following step:
> Third: Using the predicted propensity score from step first I must to rank
> households, instead of using the survey-based income alone.  Note that the
> share of the population that is eligible (all i which )is given by the
> actual program participation rate, under the counterfactual
> So this last step I can´t  to do it. Can you help me?. I have technical
> readings and literature but I can´t to make this last step with Stata. I
> suppose that I need to run a simulation, right?
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index