onearmsurv.dr

Doubly robust estimators of the coefficients in the two regression

A doubly robust precision medicine approach to fit, cross-validate and visualize prediction models for the conditional average treatment effect (CATE). It implements doubly robust estimation and semiparametric modeling approach of treatment-covariate interactions as proposed by Yadlowsky et al. (2020) <doi:10.1080/01621459.2020.1772080>.

Thomas Debray

precmed

Precision Medicine

Lu Tian

Xiaotong Jiang

Gabrielle Simoneau

Biogen  MA  Inc. 

Stan Wijn

Joana Caldas

onearmsurv.dr function

<dl><dt>ynew</dt>
<dd>Truncated survival or censoring time; vector of size <code>n</code>.</dd>
<dt>dnew</dt>
<dd>The event indicator after truncation, <code>1 = event or censored after truncation, 0 = censored before truncation</code>;
vector of size <code>n</code>.</dd>
<dt>trt</dt>
<dd>Treatment received; vector of size <code>n</code> with treatment coded as 0/1.</dd>
<dt>x.cate</dt>
<dd>Matrix of <code>p.cate</code> baseline covariates specified in the outcome model; dimension <code>n</code> by <code>p.cate</code>.</dd>
<dt>tau0</dt>
<dd>The truncation time for defining restricted mean time lost.</dd>
<dt>weightsurv</dt>
<dd>Estimated inverse probability of censoring weights with truncation for all observations; vector of size <code>n</code>.</dd>
<dt>ps</dt>
<dd>Estimated propensity scores for all observations; vector of size <code>n</code></dd>
<dt>f.predictor</dt>
<dd>Initial prediction of the outcome (restricted mean time loss) conditioned on the covariates <code>x.cate</code> for one treatment group <code>r</code>;
<code>mu_r(x.cate)</code>, step 1 in the two regression; vector of size <code>n</code></dd></dl>

Arguments

Doubly robust estimators of the coefficients in the two regression — onearmsurv.dr

<dl>

<dt>ynew</dt>
<dd>Truncated survival or censoring time; vector of size <code>n</code>.</dd>


<dt>dnew</dt>
<dd>The event indicator after truncation, <code>1 = event or censored after truncation, 0 = censored before truncation</code>;
vector of size <code>n</code>.</dd>


<dt>trt</dt>
<dd>Treatment received; vector of size <code>n</code> with treatment coded as 0/1.</dd>


<dt>x.cate</dt>
<dd>Matrix of <code>p.cate</code> baseline covariates specified in the outcome model; dimension <code>n</code> by <code>p.cate</code>.</dd>


<dt>tau0</dt>
<dd>The truncation time for defining restricted mean time lost.</dd>


<dt>weightsurv</dt>
<dd>Estimated inverse probability of censoring weights with truncation for all observations; vector of size <code>n</code>.</dd>


<dt>ps</dt>
<dd>Estimated propensity scores for all observations; vector of size <code>n</code></dd>


<dt>f.predictor</dt>
<dd>Initial prediction of the outcome (restricted mean time loss) conditioned on the covariates <code>x.cate</code> for one treatment group <code>r</code>;
<code>mu_r(x.cate)</code>, step 1 in the two regression; vector of size <code>n</code></dd>

</dl>

onearmsurv.dr: Doubly robust estimators of the coefficients in the two regression

Description

Usage

Value

Arguments