GRRGI-package {GRRGI}R Documentation

Components of Variance and their Confidence Intervals for Gauge R and R Studies

Description

Calculates components of Variance estimates for Gauge R and R Studies using ANOVA and Restricted Maximum Likelihood. Also generates the confidence intervals using the Satterthwaite approximation with the ANOVA estimates, and using the method of Generalized Inference with the maximum likelihood estimates. Also generates line plots to compare the Satterthwaite and Generalized Inference confidence limits.

Details

Package: GRRGI
Type: Package
Version: 1.1
Date: 2009-05-17
License: GPL-2
LazyLoad: yes

Gauge R and R input data need to be in a matrix with columns named 'resp', 'part' and 'operator'. Example of data sets are provided in 'HWstudy1' and 'HWstudy2'.

'anovasat' calculates the components of variance using ANOVA and the the confidence intervals for repeatability, reproducibility and total gauge using the Satterthwaite approximation.

'MLGI' calculates the components of variance using restricted maximum likelihood and the the confidence intervals using the method of generalized inference.

Version 1.1

1 Fix the labels on ratiograph. 2 Use the same random variables for both numerator and denominator of ratios (Generalized Inference). 3 Test for and stop the functions anovasat and MLGI if data is unbalanced. 4 change ML label to REML.

Author(s)

Walter Resch walt55128@msn.com

References

Package is based on Michael Hamada and Sam Weerandi (July 2000) Measurement System Assessment Via Generalized Inference, Journal of Quality Technology 32, 241-253

See Also

MLGI, anovasat, intsignif, RRgraph, ratiograph

Examples

data(HWstudy1)
sat<-anovasat(HWstudy1,0.95)
ML<-MLGI(HWstudy1,0.95,1000)
intsignif(HWstudy1)
RRgraph(sat,ML)
ratiograph(sat,ML)

[Package GRRGI version 1.1 Index]