SIsMiss: Fit the primary linear model with Nuissance-Free conditional...

In chenchi0526/SIsMiss:

Description

Fit the primary linear model with Nuissance-Free conditional likelihood method.

Usage

SIsMiss(y, z, u, regularize = FALSE, cov.names = NULL, se.method = NULL,
  CI.alpha = NULL, M = 500, seed_num = 123)

Arguments

y	Vector. The response variable subject to missing data.
z	Vector. The shadow variable (fully observed).
u	Matrix. The covariate matrix excluding shadow variable z (fully observed).
regularize	Logical. If regularize=TRUE, a regularized linear model is going to be fitted. Adaptive LASSO penalty is adopt. Default is regularize=FALSE.
cov.names	Vector. The vector of names for all the covariates. The first element is the name of variable z, followed by the name of variable u.
se.method	Charactor. The method for estimating standard error of parameters. Three options are available. "asymp" distribution. This option is applicable to regularize=FALSE only. "perturb" perturbation method. This option is applicable to both regularize=FALSE and regularize=TRUE. NULL
CI.alpha	α value for confidence interval of each parameter. CI.alpha must be a value between (0, 1). If CI.alpha=NULL, confidence interval will not be returned.
M	Number of resampling in perturbation.
seed_num	Number of seed to control randomness of perturbation term generated.

Value

A table. Summary of estimates and standard error presented in the table with p+1 rows and 2 columns.

chenchi0526/SIsMiss documentation built on Dec. 8, 2020, 2:35 a.m.