Implementing TMLE in the Presence of a Continuous Outcome
1
Citation
2
A Real data: Box 14: Initial data setup
3
Box 1: Transformation of the continuous outcome variable
4
Box 2: Fit SuperLearner & make predictions
5
Box 3a & 3b: Get predictions under both treatments,
\(A = 0\)
and
\(A = 1\)
6
Box 4: Get initial treatment effect estimate
7
Box 5: Fit propensity score SuperLearner & make predictions
8
Box 6: Estimate clever covariate
\(H\)
9
Box 7: Estimate fluctuation parameter
\(\epsilon\)
10
Box 8: Update the initial outcome model prediction based on targeted adjustment of the initial predictions using the PS model
11
Box 9: Find treatment effect estimate
12
Box 10: Transform back the treatment effect estimate in the original outcome scale
13
Box 11: Confidence interval estimation
14
Box 12: tmle package
15
Box 13: Comparison with Keele (2021)
Contact
Implementing TMLE in the Presence of a Continuous Outcome
Section 12
Box 10: Transform back the treatment effect estimate in the original outcome scale
ATE.TMLE
<-
(max.Y
-
min.Y)
*
ATE.TMLE.bounded
ATE.TMLE
## [1] 2.731947