Estimators

Inverse-probability weights (IPW)

Propensity-score matching

Covariate matching

Regression adjustment

Weighted regression

Doubly robust methods

Augmented IPW (AIPW)

IPW with regression adjustment

Watch Regression adjustment (RA).

Watch Inverse-probability weighting.

Watch Inverse-probability weights (IPW) with RA.

Watch Augmented IPW.

Watch Nearest-neighbor matching.

Watch Propensity-score matching.

 

 

Endogeneity and Heckman-style selection with treatment effects

Linear regression

Interval regression, including tobit

Probit regression

Ordered probit regression

Exogenous or endogenous regressors

Endogenous or exogenous treatment; binary or ordinal treatment

Learn about Extended regression models.

 

Statistics

Average treatment effects (ATEs)

ATEs on the treated (ATETs)

Potential-outcome means (POMs)

 

Outcomes

Continuous—linear

Binary—logistic, probit, heteroskedastic probit

Count—Poisson

Fractional

Non-negative, including exponential mean

Survival—exponential, Weibull, gamma, lognormal

Watch Treatment effects for survival models in Stata.

 

Treatments

Binary—logistic, probit, heteroskedastic probit

Multivalued-multinomial logistic

 

Diagnostics

Overlap plots

Covariate balance*

 

Endogenous treatment effects

Continuous outcome

Count outcome

Control-function estimator*

ATEs, ATETs, and POMs

Test for endogeneity

Watch Endogenous treatment effects.

 

Postestimation selector*

View and run all postestimation features for your command

Automatically updated as estimation commands are run

Watch Postestimation Selector.

Watch A tour of treatment effects.

Watch Introduction to treatment effects, part 1.

Watch Introduction to treatment effects, part 2.

 

 

Additional resource

Extended Regression Models Reference Manual

In the spotlight: Double-robust treatment effects (two wrongs don’t make a right, but one does)

In the spotlight: Treatment effects

In the spotlight: eteffects and the challenges of making causal inferences