RC: The Regression Calibration Estimator

In DylanSpicker/rcalibration: Implements Generalized Regression Calibration (and Related) Estimators for Measurement Error Correction.

Description

This function implements the standard regression calibration estimator, based on replicated measurements.

Usage

RC(W, Z = NULL, return_var = FALSE)

Arguments

W	A list of length 'k' containing matrices of error-prone proxy measurements of the covariate. Matrices should all be n (observations) x p (dimension of covariates).
Z	A matrix containing all error-free covariates for use in estimation. Matrix should be n (observations) x q (dimension). Use NULL if no such covariates exist. Defaults to NULL.
return_var	A boolean represent whether the correction function and weights should be returned (TRUE) or only the imputed values. Defaults to FALSE.

Value

Either a matrix of imputed values of size n x p (if return_var is FALSE), or a list which contains elements $X.hat (the aforementioned imputed matrix), $fitRC (a function which can be used to make the same correction), and $weights (the weights used in the correction).

See Also

[rcalibration::generalizedRC()] which this function is just a wrapper for

Examples

generalizedRC(W, weights="equal")

DylanSpicker/rcalibration documentation built on March 8, 2020, 10:38 a.m.