Compute multivariate Mahalanobis distances and corresponding chi-squared quantiles to assess multivariate normality

Compute multivariate Mahalanobis distances and corresponding chi-squared quantiles to assess multivariate normality

chisq_mvn(data)

Arguments

data	a numeric matrix or data.frame

Value

a Nx2 tibble with observed Mahalanobis distances and ordered chi-sq quantiles

Details

This function is inspired by a discussion of multivariate normality in "An R and S-Plus Companion to Multivariate Analysis" (Everitt, 2005)

Examples

mvn_coords <- chisq_mvn(iris[, -5])
#> Error in data_frame(distances = sort(di), chisq = chisq_quant): could not find function "data_frame"
plot(distances ~ chisq, data = mvn_coords)
#> Error in eval(m$data, eframe): object 'mvn_coords' not found
abline(lm(distances ~ chisq, data = mvn_coords))
#> Error in is.data.frame(data): object 'mvn_coords' not found