EMRE: EM Algorithm for Linear Regression with Missing Data

In DLMRMV: Distributed Linear Regression Models with Response Missing Variables

EM Algorithm for Linear Regression with Missing Data

Description

EM Algorithm for Linear Regression with Missing Data

Usage

EMRE(data, d = 1, tol = 1e-06, nb = 100, niter = 1)

Arguments

data	Dataframe with first column as response (Y) and others as predictors (X)
d	Initial convergence threshold (default=1)
tol	Termination tolerance (default=1e-6)
nb	Maximum iterations (default=100)
niter	Starting iteration counter (default=1)

Value

List containing:

Yhat	Imputed response vector
betahat	Estimated coefficients

Examples

# Generate data with 20% missing Y values
set.seed(123)
data <- data.frame(Y=c(rnorm(80),rep(NA,20)), X1=rnorm(100), X2=rnorm(100))

# Run EM algorithm
result <- EMRE(data, d=1, tol=1e-5, nb=50)
print(result$betahat) # View coefficients

DLMRMV documentation built on Aug. 8, 2025, 6:27 p.m.