misaem (version 1.0.1)

predict.miss.lm: Prediction on test with missing values for the linear regression model.

Description

Prediction on test with missing values for the linear regression model.

Usage

# S3 method for miss.lm
predict(object, newdata = NULL, seed = NA, ...)

Arguments

object

a fitted object of class inheriting from "miss.lm".

newdata

a data frame in which to look for variables with which to predict. It can contain missing values.

seed

An integer as a seed set for the random generator.

...

Further arguments passed to or from other methods.

Value

pr.y

The prediction result for linear regression.

Examples

# NOT RUN {



# Generate complete data
set.seed(1)
mu.X <- c(1, 1)
Sigma.X <- matrix(c(1, 1, 1, 4), nrow = 2)
n <- 50 # train set size
p <- 2 # number of covariates
X.complete <- matrix(rnorm(n*p), nrow=n)%*%chol(Sigma.X) +
              matrix(rep(mu.X,n), nrow=n, byrow = TRUE)
b <- c(2, 3, -1)
sigma.eps <- 0.25
y <- cbind(rep(1, n), X.complete) %*% b +
  rnorm(n, 0, sigma.eps)

# Add missing values
p.miss <- 0.10
patterns <- runif(n*p)<p.miss #missing completely at random
X.obs <- X.complete
X.obs[patterns] <- NA
# Estimate regression using EM
df.obs = data.frame(y ,X.obs)
miss.list = miss.lm(y~., data=df.obs)

# Generate new dataset for prediction
nt <- 20
Xt <- matrix(rnorm(nt*p), nrow=nt)%*%chol(Sigma.X)+
  matrix(rep(mu.X,nt), nrow=nt, byrow = TRUE)
# Generate missingness in new dataset
patterns <- runif(nt*p)<p.miss
Xt.obs <- Xt
Xt.obs[patterns] <- NA

# Prediction with missing values
miss.pred = predict(miss.list, data.frame(Xt.obs))
print(miss.pred)
# }