dot-wassersteinTestSp: Semi-parametric test using the 2-Wasserstein distance to...

.wassersteinTestSpR Documentation

Semi-parametric test using the 2-Wasserstein distance to check for differential distributions

Description

Two-sample test to check for differences between two distributions using the 2-Wasserstein distance: Semi-parametric implementation using a permutation test with a generalized Pareto distribution (GPD) approximation to estimate small p-values accurately

Usage

.wassersteinTestSp(x, y, permnum = 10000)

Arguments

x

sample (vector) representing the distribution of condition A

y

sample (vector) representing the distribution of condition B

permnum

number of permutations used in the permutation testing procedure

Details

This is the semi-parametric version of wasserstein.test, for the asymptotic theory-based procedure see .wassersteinTestAsy.

Details concerning the permutation testing procedure with GPD approximation to estimate small p-values accurately can be found in Schefzik et al. (2020).

Value

A vector of 15, see Schefzik et al. (2020) for details:

  • d.wass: 2-Wasserstein distance between the two samples computed by quantile approximation

  • d.wass^2: squared 2-Wasserstein distance between the two samples computed by quantile approximation

  • d.comp^2: squared 2-Wasserstein distance between the two samples computed by decomposition approximation

  • d.comp: 2-Wasserstein distance between the two samples computed by decomposition approximation

  • location: location term in the decomposition of the squared 2-Wasserstein distance between the two samples

  • size: size term in the decomposition of the squared 2-Wasserstein distance between the two samples

  • shape: shape term in the decomposition of the squared 2-Wasserstein distance between the two samples

  • rho: correlation coefficient in the quantile-quantile plot

  • pval: p-value of the semi-parametric 2-Wasserstein distance-based test

  • p.ad.gpd: in case the GPD fitting is performed: p-value of the Anderson-Darling test to check whether the GPD actually fits the data well (otherwise NA).

  • N.exc: in case the GPD fitting is performed: number of exceedances (starting with 250 and iteratively decreased by 10 if necessary) that are required to obtain a good GPD fit, i.e. p-value of Anderson-Darling test \geq 0.05 (otherwise NA).

  • perc.loc: fraction (in %) of the location part with respect to the overall squared 2-Wasserstein distance obtained by the decomposition approximation

  • perc.size: fraction (in %) of the size part with respect to the overall squared 2-Wasserstein distance obtained by the decomposition approximation

  • perc.shape: fraction (in %) of the shape part with respect to the overall squared 2-Wasserstein distance obtained by the decomposition approximation

  • decomp.error: relative error between the squared 2-Wasserstein distance obtained by the quantile approximation and the squared 2-Wasserstein distance obtained by the decomposition approximation

References

Schefzik, R., Flesch, J., and Goncalves, A. (2020). waddR: Using the 2-Wasserstein distance to identify differences between distributions in two-sample testing, with application to single-cell RNA-sequencing data.


goncalves-lab/diffexpR documentation built on June 5, 2023, 10:18 p.m.