acf.calc: Autocorrelation function

In hli226/mvnimpute: Simultaneously Impute the Missing and Censored Values

Autocorrelation function

Description

Calculates the autocorrelation function and draws the plots.

Usage

acf.calc(data.mat, lag = 50, plot = TRUE, title = NULL, details = FALSE)

Arguments

data.mat	matrix including the variables of which autocorrelations are calculated.
lag	lag at which the autocorrelation is calculated, default is set as 50.
plot	logical variable to specify whether the plot is generated, default is set to TRUE.
title	title of each generated autocorrelation plot.
details	boolean variable to specify whether the autocorrelation values are returned, default is set to FALSE.

Details

This function calculates the autocorrelations of all the variables on a column by column base. The default value of lag is set as 50, the maximum number of lag should not exceed the number of rows of the dataset, which reflects the corresponding number of iteration of running the multiple imputation.

Value

If details = TRUE, a matrix containing the calculated autocorrelations of all the variables in the dataset will be returned. If plot = TRUE, the autocorrelation plots of all the variables will be drawn.

Examples

### generate some data
dat <- MASS::mvrnorm(n = 1000, mu = c(1, 2, 3, 4), Sigma = diag(4))

### ACF plots
acf.calc(data.mat = dat, title = paste0("Var ", 1:nrow(dat)))

hli226/mvnimpute documentation built on July 6, 2022, 11:14 a.m.