blipVarianceTMLE: Compute TMLE on the variance of CATE (binary Y)

In wilsoncai1992/TMLEbootstrap: Bootstrap Confidence Intervals for Targeted Maximum Likelihood Estimators

Description

Compute TMLE on the variance of CATE (binary Y)

Compute TMLE on the variance of CATE (binary Y)

Details

inherit the initial fit routine from ate

Super class

TMLEbootstrap::ateTMLE -> blipVarianceTMLE

Methods

Public methods

• blipVarianceTMLE$initial_fit()

• blipVarianceTMLE$target()

• blipVarianceTMLE$compute_EIC()

• blipVarianceTMLE$inference_without_target()

• blipVarianceTMLE$compute_min_phi_ratio()

• blipVarianceTMLE$clone()

Inherited methods

• TMLEbootstrap::ateTMLE$initial_fit_pen_likeli()

• TMLEbootstrap::ateTMLE$initialize()

• TMLEbootstrap::ateTMLE$plot_Q1W()

---

Method initial_fit()

Usage

blipVarianceTMLE$initial_fit(family_y = "binomial", ...)

---

Method target()

use the logistic submodel to target blip variance; output Psi, EIC, CI

Usage

blipVarianceTMLE$target()

Returns

NULL

---

Method compute_EIC()

Compute EIC of blip variance

Usage

blipVarianceTMLE$compute_EIC(Y, A, Q1k, Q0k, Qk, gk, psi)

Arguments

Y

continuous target variable

A

binary treatment

Q1k

E(Y|A=1,W)

Q0k

E(Y|A=0,W)

Qk

E(Y|A=a,W)

gk

E(A=a|W)

psi

var(E(Y|A=1,W) - E(Y|A=0,W))

Returns

EIC vector

---

Method inference_without_target()

Usage

blipVarianceTMLE$inference_without_target()

---

Method compute_min_phi_ratio()

Usage

blipVarianceTMLE$compute_min_phi_ratio()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

blipVarianceTMLE$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

wilsoncai1992/TMLEbootstrap documentation built on Feb. 27, 2021, 7:57 a.m.