SurrogateRegression: Surrogate Outcome Regression Analysis

Performs estimation and inference on a partially missing target outcome while borrowing information from a correlated surrogate outcome to increase estimation precision and improve power. The target and surrogate outcomes are jointly modeled within a bivariate outcome regression framework. Unobserved values of either outcome are regarded as missing data. Estimation in the presence of bilateral outcome missingness is performed via an expectation conditional maximization algorithm. A flexible association test is provided for evaluating hypotheses about the target regression parameters. See McCaw ZR, Gaynor SM, Sun R, Lin X; “Cross-tissue eQTL mapping in the presence of missing data via surrogate outcome analysis” <doi:10.1101/2020.11.29.403063>.

Version: 0.5.0
Depends: R (≥ 3.4.0)
Imports: methods, mvnfast, plyr, Rcpp, stats
LinkingTo: Rcpp, RcppArmadillo
Suggests: knitr, rmarkdown
Published: 2020-12-03
Author: Zachary McCaw ORCID iD [aut, cre]
Maintainer: Zachary McCaw <zmccaw at alumni.harvard.edu>
License: GPL-3
NeedsCompilation: yes
CRAN checks: SurrogateRegression results

Downloads:

Reference manual: SurrogateRegression.pdf
Vignettes: Surrogate Outcome Regression Analysis
Package source: SurrogateRegression_0.5.0.tar.gz
Windows binaries: r-devel: SurrogateRegression_0.5.0.zip, r-release: SurrogateRegression_0.5.0.zip, r-oldrel: SurrogateRegression_0.5.0.zip
macOS binaries: r-release: SurrogateRegression_0.5.0.tgz, r-oldrel: SurrogateRegression_0.5.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=SurrogateRegression to link to this page.