AirScreen (version 0.1.0)

rnormAR1: Generate normal samples (Autoregressive AR(1) correlation)

Description

rnormAR1 efficiently generates samples from a multivariate normal distribution with AR(1) correlation \(cor(x_i, x_j) = \rho^{\lvert i-j \rvert}\).

Usage

rnormAR1(n, p, rho = 0.5, means = 0, variances = 1)

Value

A numeric \(n \times p\) matrix with the specified correlation, means, and variances.

Arguments

n

Number of observations.

p

Number of variables.

rho

Autoregressive correlation coefficient.

means

Numeric vector of feature means (length \(1\) or \(p\)).

variances

Numeric vector of feature variances (length \(1\) or \(p\)).

Examples

X1 <- rnormAR1(10, 5)
X2 <- rnormAR1(10, 5, rho = 0.3, means = 2, variances = 4)
X3 <- rnormAR1(10, 5, rho = 0.4, means = 1:5, variances = 3:7)