disagreement between p value and confidence intervals using stteffects ipwra

[Mazen Zenati]

Hello
I am using "stteffects ipwra" and in the output, the ATE ginving coef. = 4.9 std., Err. = 2.1, z= 2.3 p=0.021, and 95% conf. interval 0.751, 9.114. Anyone can tell way such contradiction between p value and 95% confidence interval?

Note: The second line (Pomean) provided with p= 0.000 and 95% C.I =16.8, 22.72 (no contradiction)

Stata question.docx

[William Lisowski]

Welcome to Statalist.

I do not see the contradiction. The 95% confidence interval does not include 0, so that tells us that the p-value for a two-sided test of the null hypothesis that the coefficient is zero must be less than 0.05. The p-value show of 0.021 is indeed less than 0.05.

I note also that (9.114-4.932)/2.133 = 1.98, which is what one would expect for a 95% confidence interval based on a coefficient estimate of 9.114 and a standard error of 2.133.