Chi-Plots for Assessing Bivariate and Multivariate Dependence

Graphical methods for assessing dependence. The chi-plot of Fisher and Switzer (1985) and (2001) supplements a scatterplot of two variables, producing characteristic patterns for different forms of association. Its multivariate extension (Fisher and Lee, 2026) uses generalized orthants and data depth to explore whether a multivariate distribution factorizes into independent blocks of variables, with permutation tests and guidelines. Functions are provided for single chi-plots, a scatterplot matrix with chi-plots above the diagonal, multivariate chi-plots for a given partition of the variables, and a search over all partitions. Computation-intensive parts are written in C.