Kpart

Description

Kpart spline fitting

Details

~~ This package is intended for use with non-lineraly associated data. The function lm.Kpart firsts selects points for cubic spline knots using an alogorithm to find the abolute maximum deviate from the partiton mean, then fits a best fitting model by using the best subset method and minimum BIC. The function knots.Kpart returns the values selcted as knots in the model given by lm.Kpart, and plot.Kpart returns a plot with fitted values and vertical lines representing the spline points. Kpart(data,K) takes two arguments, data is a two column matrix where the first column is the repsonse and the second column is the predictor upon which knots are selected. K is a postive interger that indicates how many equally spaced partitions the user would like to produce.~~

Author(s)

Eric Golinko

Maintainer: egolinko@gmail.com

Results