Performs a permutation test of differences in means between groups of censored data.
cenpermanova(y1, y2, grp, R = 9999, printstat = TRUE)The column of data values plus detection limits.
The column of indicators, where 1 (or TRUE) indicates a detection limit in the y1 column, and 0 (or FALSE) indicates a detected value in y1.
Grouping or factor variable. Can be either a text or numeric value indicating the group assignment.
The number of permutations used. Default is 9999.
Logical TRUE/FALSE option of whether to print the resulting statistics in the console window, or not. Default is TRUE.
Permutation test results with the number of permutations, range in test statistics and p-value values through the various permutations. Group means are also listed.
Because this is a permutation test it avoids the problem with MLE tests (see cenanova) that assume a normal distribution. No values are modeled as below zero and group means and p-values are trustworthy.
Good, P., 2000. Permutation Tests: A Practical Guide to Resampling Methods for Testing Hypotheses, 2nd ed, Springer Series in Statistics. Springer-Verlag, New York, NY. doi: 10.1007/978-1-4757-3235-1
Helsel, D.R., 2011. Statistics for Censored Environmental Data using Minitab and R, 2nd ed. John Wiley & Sons, USA, N.J.
data(PbHeron)
cenpermanova(PbHeron$Liver,PbHeron$LiverCen,PbHeron$DosageGroup)
#> Permutation test of mean CensData: PbHeron$Liver by Factor: PbHeron$DosageGroup
#> 9999 Permutations
#> Test Statistic = 1122 to 1122 p = 0.0022 to 0.0023
#>
#> mean(High) mean(Low)
#> 9.2880 0.1204