sample.knowledge.H1: Draw a random sample from assumed distribution of H1 values....

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

Description

A sample of a certain size is drawn from a chosen distribution. The parameters for the distribution are obtained through a call to the function par.dist.default. The parameters may be changed, although the structure should not be changed.

For alternative hypothesis H1C one should use the truncated normal distribution for the parameter ”par”.

Usage

sample.knowledge.H1(n, par, p.0=NULL)

Arguments

n	The size of the sample to be drawn
par	The distribution and its parameters.
p.0	A vector with the PD values under H0. The default value is NULLL. If a vector is supplied, then, for each draw, only the p.1 rows where at least one pd is higher than the corresponding p.0 pd are kept.

Details

A sample of size n is drawn from the distribution.

Value

A list with elements:

dist	the name of the distribution that was used
par	the parameters of the distribution that was used
p.1	the sample drawn
stats.wrong.pds	The p.1 are randomly drawn from the distribution given in the parameter par. By the random drawing it might be that the outcome does not fulfill H1. In that case the draw is excluded from the list-element p.1 and added to the list element stats.wrong.pds

See Also

par.dist.default, simul.scenario.rs

Examples

## Not run: 
s.nrm<-sample.knowledge.H1(n=100, par=par.dist.default(dist="tr.normal", 
                         p.0=c(0.001, 0.004)))

## End(Not run)

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