STATA - Mean differences between treated and control groups "after matching"

[Jewon Shin]

1. Relevant Files: https://www.dropbox.com/sh/0jnj3txf4...qG2p5emwa?dl=0

import excel "merge_db_environment_korea_kospi200_findel.xl sx", firstrow
keep if ksic1 != "K"
gen CO2Sale = co2_sale_gr_1000
keep if year<=2019
keep if year>=2011

global control_lag = "lag_size lag_cf_vol lag_capex lag_leverage lag_roa lag_ln_firm_age lag_rnd_sale"
global ESG_lag "lag_E_num lag_G_num lag_S_num"
global d_control_lag = "d_size d_cf_vol d_capex d_leverage d_roa d_ln_firm_age d_rnd_sale"
global d_ESG_lag "d_E_num d_G_num d_S_num"
egen ksic1_num = group(ksic1), label

teffects nnmatch (CO2Sale ${control_lag} ${ESG_lag} ${d_control_lag} ${d_ESG_lag} i.ksic1_num i.year) (CarbonOffset), gen(match)



2. I want to conduct a "t-difference mean test between treated and control groups after matching".

For instance, I can calculate a "t-difference mean test between treated and control groups" before matching like this:
ttest CO2Sale,by(CarbonOffset) level(99) unequal

And I can also calculate an Average Treatment Effect (ATE) like this as well:
teffects nnmatch (CO2Sale ${control_lag} ${ESG_lag} ${d_control_lag} ${d_ESG_lag} i.ksic1_num i.year) (CarbonOffset), gen(match)


However, I don't know how to conduct a "t-difference mean test between treated and control groups after matching". If there is someone who can conduct a similar test, I really want to know how to do it.

[Jared Greathouse]

webuse cattaneo2, clear
cls
teffects nnmatch (bweight mage prenatal1 mmarried fbaby) (mbsmoke)
tebalance summ

About right?


EDIT: You mentioned a t-test, psmatch2 will do this for you (I believe it's still on ssc)