Checking estimation consistency

nhejazi / txshift

:package: :game_die: R/txshift: Efficient Estimation of the Causal Effects of Stochastic Interventions, with Corrections for Outcome-Dependent Sampling

https://codex.nimahejazi.org/txshift

Other

13 stars 4 forks source link

Checking estimation consistency #1

Closed nhejazi closed 6 years ago

nhejazi commented 6 years ago

For density estimation with condensier:

For outcome estimation with Super Learner (via sl3):
- Predict on a new data set for different values of A (e.g. from –4 to 4) and with W = 0.
- Make a scatter plot of Y[W = 0]. Add the prediction line. Add the true line.
- Repeat the above for W = 1.

nhejazi commented 6 years ago

These issues are mostly addressed by the simple simulation studies implemented in the various .Rmd files provided as of bdef728.