Hi all -
I am applying the command from Emsley, Lunt, Pickle and Dunn (The State Journal 2018 https://journals.sagepub.com/doi/pdf...867X0800800302). "dr" command seems to work fine but the issue is independent variables, sample size as well as R-squared are not tabulated, only the "exposure" or the variable of interest/indicator is shown in the output. However, there are published papers (across all disciplines) that use this method show the complete set of results. So I wonder if I miss something or Stata per se is unable to display them. See the below example:
Only a is presented no matter which model is used.
Model 1: dr y a
Doubly Robust Estimate of the effect of a on y
Using sandwich estimator of SE
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
a | .3269012 .0862426 3.79 0.000 .1578689 .4959335
------------------------------------------------------------------------------
Model 2: dr y a ybase
Doubly Robust Estimate of the effect of a on y
Using sandwich estimator of SE
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
a | .3002054 .0785934 3.82 0.000 .1461652 .4542455
------------------------------------------------------------------------------
Model 3: dr y a ybase x
Doubly Robust Estimate of the effect of a on y
Using sandwich estimator of SE
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
a | .4826762 .0885831 5.45 0.000 .3090566 .6562958
------------------------------------------------------------------------------
Model 4: dr y a, ovars(ybase x intxy ysqr intxy2) pvars(x ybase intxy)
Doubly Robust Estimate of the effect of a on y
Using sandwich estimator of SE
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
a | .4243645 .0699094 6.07 0.000 .2873447 .5613843
------------------------------------------------------------------------------
Thanks much.
Related Posts with Double - robust regression estimator in Stata
Time Series and Fixed Effects Estimation ProblemsI am a beginner in Stata and I am having some trouble when undertaking regressions in Stata. For so…
r(134) too many valuesDear Statalist, I've run my regressions and now I would like to confirm my results on a panel of fi…
Hausman test - gives p value as dot, not even 0.00Please advice why Prob>chi2 is given as a . in my test, rather than a value for me to interpret. …
When to reject H0 fisher-type unit root testWhat to do when the P-value of chi-squared is significant but normal inverse Z is not? …
upside down bar graphHi, Can someone please tell what code in STATA can I use to make this upside down bar chart. I coul…
Subscribe to:
Post Comments (Atom)
0 Response to Double - robust regression estimator in Stata
Post a Comment