# [R] [MatchIt] Naive Estimator for ATT after Full Matching

thebudget72 m@iii@g oii gm@ii@com thebudget72 m@iii@g oii gm@ii@com
Wed May 5 20:55:09 CEST 2021

```Dear R-help ML,

I would like to compute a Naive Estimator for the Average Treatment
Effect (ATT) after a Propensity Score Matching with full matching.

Since it is full matching, the resulting post-matching database contains
all the observations of the original dataset.

I came up with this code, which does a weighted average of the outcomes,
using the weights provided by the matching process, but I'm not sure
this is the correct way to achieve it.

How can I compute the ATT using a Naive Estimator after PSM?

I know I am supposed to do a regression, but I am interested in
computing a Naive Estimator as a difference between the means across the
two groups.

```r
library("MatchIt")
data("lalonde")

m.out2 <- matchit(treat ~ age + educ + race + married +
nodegree + re74 + re75,
data = lalonde,
method = "full",
distance = "glm",

m.data2 <- match.data(m.out2)

te <- weighted.mean(m.data2\$re78[m.data2\$treat],
m.data2\$weights[m.data2\$treat])
nte <- weighted.mean(m.data2\$re78[!m.data2\$treat],
m.data2\$weights[!m.data2\$treat])
ne2w <- round(te-nte, 2)

print(paste0("The ATT estimated with a NE is: ", ne2w))
```

Thanks in advance and best regards.

```