Six sigma
GenStat has wide range of facilities to support the six-sigma approach to quality improvement.
It can display many different types of control chart.
plots c or u charts representing numbers of defective items
prints CUSUM tables for controlling a process mean
plots exponentially weighted moving-average control charts
plots p or np charts for binomial testing for defective items
plots control charts for mean and standard deviation or
range
It can test for Normality, display Pareto charts and calculate capability statistics.
performs tests of univariate and/or multivariate normality
calculates capability statistics
sorts tables to put margins are in ascending or descending order
for display as a Pareto chart
It also provides full statistical backup for wider-ranging investigations. The list below highlights
some of the commands that may be useful.
uses the BLKL algorithm to construct response-surface designs
generates Box-Behnken designs
generates central composite designs
selects from a set of standard designs including factorials with
interactions confounded with blocks
generates fractional factorial designs
generates designs to estimate main effects of two-level
factors (Plackett-Burman designs)
plots one- or two-way tables of means from
ANOVA
plots residuals from an
ANOVA analysis
performs pairwise multiple comparison tests for
ANOVA means
performs analysis of variance for unbalanced designs
fits a linear, generalized linear, generalized additive, or generalized
nonlinear model
fits a standard nonlinear regression model
fits a nonlinear regression model or optimizes a
function
forms design keys for balanced designs with several error terms,
allowing for confounded and aliased treatments
fits an unbalanced linear mixed model and estimates variance
components
fits a quadratic surface and estimates its stationary point
estimates the parameter lambda from various single-parameter transformations, includling power (Box-Cox), modulus, folded power, Guerrero-Johnson, Aranda-Ordaz and power logit