Power Simulation for One-Factorial All-Pairs and Many-To-One Comparison Tests
Performs power simulation for one-factorial all-pairs and Many-To-One comparison tests.
powerMCTests(
mu,
n = 10,
errfn = c("Normal", "Lognormal", "Exponential", "Chisquare", "TDist", "Cauchy",
"Weibull"),
parms = list(mean = 0, sd = 1),
test = c("kwManyOneConoverTest", "kwManyOneDunnTest", "kwManyOneNdwTest",
"vanWaerdenManyOneTest", "normalScoresManyOneTest", "dunnettTest",
"tamhaneDunnettTest", "ManyOneUTest", "kwAllPairsNemenyiTest", "kwAllPairsDunnTest",
"kwAllPairsConoverTest", "normalScoresAllPairsTest", "vanWaerdenAllPairsTest",
"dscfAllPairsTest", "gamesHowellTest", "lsdTest", "scheffeTest", "tamhaneT2Test",
"tukeyTest", "dunnettT3Test", "pairwise.t.test", "pairwise.wilcox.test",
"adManyOneTest", "adAllPairsTest", "bwsManyOneTest", "bwsAllPairsTest",
"welchManyOneTTest"),
alternative = c("two.sided", "greater", "less"),
p.adjust.method = c("single-step", p.adjust.methods),
alpha = 0.05,
FWER = TRUE,
replicates = 1000
)mu |
numeric vector of group means. |
n |
number of replicates per group. If |
errfn |
the error function. Defaults to |
parms |
a list that denotes the arguments for the error function.
Defaults to |
test |
the multiple comparison test for which the power analysis is
to be performed. Defaults to |
alternative |
the alternative hypothesis. Defaults to |
p.adjust.method |
method for adjusting p values (see |
alpha |
the nominal level of Type I Error. |
FWER |
logical, indicates whether the family-wise error should be computed.
Defaults to |
replicates |
the number of Monte Carlo replicates or runs. Defaults to |
The linear model of a one-way ANOVA can be written as:
X_{ij} = μ_i + ε_{ij}
For each Monte Carlo run, the function simulates ε_{ij} based on the given error function and the corresponding parameters. Then the specified all-pairs or many-to-one comparison test is performed. Finally, several effect sizes (Cohen's f ans R-squared), error rates (per comparison error rate, false discovery rate and familywise error rate) and test powers (any-pair power, average per-pair power and all-pairs power) are calculated.
An object with class powerPMCMR.
## Not run: mu <- c(0, 0, 1, 2) n <- c(5, 4, 5, 5) set.seed(100) powerMCTests(mu, n, errfn="Normal", parms=list(mean=0, sd=1), test="dunnettTest", replicates=1E4) powerMCTests(mu, n, errfn="Normal", parms=list(mean=0, sd=1), test="kwManyOneDunnTest", p.adjust.method = "bonferroni", replicates=1E4) ## End(Not run)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.