simul.scenario.rs: Determine cardinality and power for a statistical test.

In validateRS: One-Sided Multivariate Testing Procedures for Rating Systems

Description

For a sample drawn with sample.knowledge.H1 and for a statistical test, the acceptance region is determined and the power for each value of H1 in the sample is computed.

Usage

simul.scenario.rs(hypo.test, p.0, sampleH1, sizes, alpha)

Arguments

hypo.test	The name of the hypothesis test: “minP” for the multiple test , “minPp” for the enhanced multiple test or “sterneHull” for the envelope test.
p.0	A vector containing the PDs under H0
sampleH1	The sample drawn using the function sample.knowledge.H1.
sizes	A matrix with different scenario's for the sizes of the rating classes. There should be at least two rows and the number of columns must equal the length of the parameter p.0.
alpha	The significance level

Value

A list with elements

par	The parameters passed to the function
power	Power for each value of H1, the average power and the standard error on the average
cardinality.AR	The cardinality of the acceptance region
executionTime	The time it took to compute the acceptance region and to compute the power
knowledgeH1	The characteristics ot the distribution and the sample drawn

See Also

sample.knowledge.H1,

Examples

p.0<-c(0.001, 0.004)
sizes<-rbind( c(500, 500), 
              c(1000, 5000))
alpha<-0.05
## Not run: 
s.nrm<-sample.knowledge.H1(n=10, 
                        par=par.dist.default(dist="tr.normal", p.0=c(0.001, 0.004)))
scen.nrm.mp<-simul.scenario.rs(hypo.test="minP", p.0=p.0, 
                        sampleH1=s.nrm, sizes=sizes, alpha=alpha)
scen.nrm.mpp<-simul.scenario.rs(hypo.test="minPp", p.0=p.0, 
                        sampleH1=s.nrm, sizes=sizes, alpha=alpha)

par.ownsample<-par.dist.default(dist="usersupplied", p.0=c(0.001, 0.004))
par.ownsample$param<-rbind(c(0.001, 0.01), c(0.004, 0.004), c(0.004, 0.01))
s.own<-sample.knowledge.H1(n=10, 
                        par=par.ownsample)

scen.ownsample.mp<-simul.scenario.rs(hypo.test="minP", p.0=p.0, 
                        sampleH1=s.own, sizes=sizes, alpha=alpha)

## End(Not run)

validateRS documentation built on May 2, 2019, 3:47 p.m.