IAcsSPCR (version 1.2.1)

MEWMA: Multivariate EWMA Control Chart

Description

Computes a MEWMA using the method of Lowry, Woodall, Champ and Rigdon. The number of variables p must be between 2 and 10, r is fixed at .1

Usage

MEWMA(X,Sigma=NULL,mu=NULL,Sigma.known=TRUE)

Arguments

X

input - this is a matrix or data frame containing the multivariate data. One line for each observation and one column for each variable or quality characteristic being monitored.

Sigma

input this is the known (or estimate from a Phase I study) covariance matrix of the variables

mu

input this is the known (or estimate from a Phase I study) mean vector of the variables

Sigma.known

input this is a logical variable, if TRUE, Sigma, and mu must be supplied, if FALSE the function will estimate them from the data in X

Value

returned list containing the upper control limit, the covariance matrix and the mean vector.

References

Lowry, Woodall, Champ and Rigdon(1992)<https://www.tandfonline.com/doi/abs/10.1080/00401706.1992.10485232.>

Examples

# NOT RUN {
data(Lowry)
Sigma<-matrix(c(1, .5, .5, 1), nrow=2, ncol=2)
mu<-c(0,0)
MEWMA(Lowry,Sigma,mu,Sigma.known=TRUE)
MEWMA(Lowry,Sigma.known=FALSE)
mu5<-c(-.314,.32)
Sig5<-matrix(c(1.16893, -.3243, -.3243, 1.16893), nrow=2, ncol=2)
MEWMA(Lowry,Sig5,mu5,Sigma.known=TRUE)
# }