predict: Predictions from a harvested tree

In Harvest.Tree: Harvest the Classification Tree

Description

The function predict computes the prediction of membership from a new data set classified by harvested classification model of training data.

Usage

predict(harfunc.object, data, num.var)

Arguments

harfunc.object	the output of harfunc function.
data	test data
num.var	number of explaining varibles

Details

To run the predict function, a trained harvested classification tree formed by harvest function is required.

Value

pred.mat is a data frame stored the information of result of prediction with the following columns:

belong the node that data point belongs to

possibility the probability of point being in class 1

predict the simple perdict based on whether probability is larger than 0.5.

Harvest.Tree documentation built on May 2, 2019, 3:31 p.m.