Hi all,
I am using kmatch at stata 16.0. The command "kmatch md publiccomp ta poso_015 (apopbtdas), nate att nn //
lincom ATT-NATE" shows the output below:
kmatch md publiccomp ta poso_015 (apopbtdas), nate att nn
Multivariate-distance nearest-neighbor matching
Number of obs = 707,170
Neighbors: min = 1
Treatment : publiccomp = 1 max = 50
Metric : mahalanobis
Covariates : ta poso_015
Matching statistics
------------------------------------------------------------------------------
| Matched | Controls
| Yes No Total | Used Unused Total
-----------+---------------------------------+--------------------------------
Treated | 218644 0 218644 | 141726 346800 488526
------------------------------------------------------------------------------
Treatment-effects estimation
------------------------------------------------------------------------------
apopbtdas | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ATT | .3042954 .2101507 1.45 0.148 -.1075931 .7161839
NATE | .0177991 .2272304 0.08 0.938 -.4275651 .4631634
------------------------------------------------------------------------------
( 1) ATT - NATE = 0
------------------------------------------------------------------------------
apopbtdas | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | .2864962 .0939816 3.05 0.002 .1022954 .4706971
------------------------------------------------------------------------------
Therefore, can I conclude that the "effect" of the variable publiccomp on variable apopbtdas is statistically significant according to t-stat (3.05) ?
Thanks,
Angeliki
I am using kmatch at stata 16.0. The command "kmatch md publiccomp ta poso_015 (apopbtdas), nate att nn //
lincom ATT-NATE" shows the output below:
kmatch md publiccomp ta poso_015 (apopbtdas), nate att nn
Multivariate-distance nearest-neighbor matching
Number of obs = 707,170
Neighbors: min = 1
Treatment : publiccomp = 1 max = 50
Metric : mahalanobis
Covariates : ta poso_015
Matching statistics
------------------------------------------------------------------------------
| Matched | Controls
| Yes No Total | Used Unused Total
-----------+---------------------------------+--------------------------------
Treated | 218644 0 218644 | 141726 346800 488526
------------------------------------------------------------------------------
Treatment-effects estimation
------------------------------------------------------------------------------
apopbtdas | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ATT | .3042954 .2101507 1.45 0.148 -.1075931 .7161839
NATE | .0177991 .2272304 0.08 0.938 -.4275651 .4631634
------------------------------------------------------------------------------
( 1) ATT - NATE = 0
------------------------------------------------------------------------------
apopbtdas | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | .2864962 .0939816 3.05 0.002 .1022954 .4706971
------------------------------------------------------------------------------
Therefore, can I conclude that the "effect" of the variable publiccomp on variable apopbtdas is statistically significant according to t-stat (3.05) ?
Thanks,
Angeliki
