ks.srm.test: KS test for software reliability models

In SwReliab/gof4srm: Goodness-of-fit tests for NHPP-based software reliability models

Description

Perform the KS test for the estimated software reliability models

Usage

ks.srm.test(
  obj,
  alternative = c("two.sided", "less", "greater"),
  method = c("mc", "gks", "mc2", "yamada"),
  ...
)

Arguments

obj	An object of an estimated result by Rsrat
alternative	A string indicates the alternative hypothesis and must be one of "two.sided" (default), "less", or "greater".
method	A string indicates the method for KS test with group data. "mc" and "mc2" are Monte-Carlo methods. "gks" is the generalized KS test. "yamada" is an approximation. The default is "mc".
...	A list of other options that are passed to concrete functions.

Value

A list with components;

statistic	A value of the test statistic.
p.value	A value of the p-value of the test.
p.value.lower	A lower bound of the p-value of the test. (gks)
p.value.upper	A upper bound of the p-value of the test. (gks)
alternative	A string of the alternative hypothesis.
b	An integer for the number of resamples. (mc, mc2)
alpha	A value of the significant level. (mc, mc2)
cv	A value of the coefficient variation. (mc, mc2)
lhc	A logical indicates resamples are drawn by Latin hypercude. (mc, mc2)
seed	An integer for the seed of random numbers. (mc, mc2)

Examples

data(dacs)
result <- fit.srm.nhpp(time=sys1[sys1>=0],
                       te=-sys1[sys1<0], srm.name=c("exp"))
ks.srm.test(result)
results <- fit.srm.nhpp(time=sys1[sys1>=0],
                        te=-sys1[sys1<0], srm.name=c("exp", "gamma"))
sapply(results, function(x) x$p.value)

SwReliab/gof4srm documentation built on Dec. 18, 2021, 3:05 p.m.