Matrix Tests
A package dedicated to running multiple statistical hypothesis tests on rows and columns of matrices.
Goals
- Fast execution via vectorization.
- Convenient and detailed output format.
- Compatibility with tests implemented in base R.
- Careful handling of missing values and edge cases.
Examples
1. Bartlett's test on columns
Bartlett's test on every column of iris dataset using Species as groups:
col_bartlett(iris[,-5], iris$Species) obs.tot obs.groups var.pooled df statistic pvalue
Sepal.Length 150 3 0.26500816 2 16.005702 0.0003345076070163084
Sepal.Width 150 3 0.11538776 2 2.091075 0.3515028004158132768
Petal.Length 150 3 0.18518776 2 55.422503 0.0000000000009229038
Petal.Width 150 3 0.04188163 2 39.213114 0.0000000030547839322
2. Welch t-test on rows
Welch t-test performed on each row of 2 large (million row) matrices:
X <- matrix(rnorm(10000000), ncol = 10)
Y <- matrix(rnorm(10000000), ncol = 10)
row_t_welch(X, Y) # running time: 2.4 secondsConfidence interval computations can be turned-off for further increase in speed:
row_t_welch(X, Y, conf.level = NA) # running time: 1 secondAvailable Tests
| Variant | Name | Function |
|---|---|---|
| Location tests (1 group) | Single sample Student's t.test | row_t_onesample(x) |
| Single sample Wilcoxon's test | row_wilcoxon_onesample(x) |
|
| Location tests (2 groups) | Equal variance Student's t.test | row_t_equalvar(x, y) |
| Welch adjusted Student's t.test | row_t_welch(x, y) |
|
| Two sample Wilcoxon's test | row_wilcoxon_twosample(x, y) |
|
| Location tests (paired) | Paired Student's t.test | row_t_paired(x, y) |
| Paired Wilcoxon's test | row_wilcoxon_paired(x, y) |
|
| Location tests (2+ groups) | Equal variance oneway anova | row_oneway_equalvar(x, g) |
| Welch's oneway anova | row_oneway_welch(x, g) |
|
| Kruskal-Wallis test | row_kruskalwallis(x, g) |
|
| van der Waerden's test | row_waerden(x, g) |
|
| Scale tests (2 groups) | F variance test | row_f_var(x, y) |
| Scale tests (2+ groups) | Bartlett's test | row_bartlett(x, g) |
| Fligner-Killeen test | row_flignerkilleen(x, g) |
|
| Levene's test | row_levene(x, g) |
|
| Brown-Forsythe test | row_brownforsythe(x, g) |
|
| Association tests | Pearson's correlation test | row_cor_pearson(x, y) |
| Periodicity tests | Cosinor | row_cosinor(x, t, period) |
| Distribution tests | Jarque-Bera test | row_jarquebera(x) |
| Anderson-Darling test | row_andersondarling(x) |
Further Information
For more information please refer to the Wiki page:
See Also
Literature
Computing thousands of test statistics simultaneously in R, Holger Schwender, Tina Müller.
Statistical Computing & Graphics. Volume 18, No 1, June 2007.
Packages
CRAN:
-
ttests()in the Rfast package. -
row.ttest.stat()in the metaMA package. -
MultiTtest()in the ClassComparison package. -
bartlettTests()in the heplots package. -
harmonic.regression()in the HarmonicRegression package.
BioConductor:
-
lmFit()in the limma package. -
rowttests()in the genefilter package. -
mt.teststat()in the multtest package. -
row.T.test()in the HybridMTest package. -
rowTtest()in the viper package. -
lmPerGene()in the GSEAlm package.
GitHub:
