Greg McMahan

Greg McMahan

3 packages on CRAN

dann

cran
99.99th

Percentile

Discriminant Adaptive Nearest Neighbor Classification is a variation of k nearest neighbors where the shape of the neighborhood is data driven. This package implements dann and sub_dann from Hastie (1995) <https://web.stanford.edu/~hastie/Papers/dann_IEEE.pdf>.

99.99th

Percentile

Creates family objects identical to stats family but for new links.

99.99th

Percentile

Have you ever struggled to find "good data" for a generalized linear model? Would you like to test how quickly statistics converge to parameters, or learn how picking different link functions affects model performance? This package creates ideal data for both common and novel generalized linear models so your questions can be empirically answered.