cv.lo2: Cross-validating Loess curve

In davidaarmstrong/damisc_nodep: Dave Armstrong's Miscellaneous Functions

Description

Function provides the cross-validation error for the loess curve in a manner that is amenable to optimization of the span.

Usage

cv.lo2(span, form, data, cost = function(y, yhat) mean((y - yhat)^2, na.rm=T), 
    K = n, numiter = 100, which=c("corrected", "raw"))

Arguments

span	The span of the loess smoother.
form	The formula that identifies the model
data	A data frame containing the required variables.
cost	Cost function to be passed down to loess.
K	Number of folds for the cross-validation
numiter	Number of times over which the cv error will be aggregated
which	Return raw or corrected cv error

Value

The cross-validation error from the loess curve.

Author(s)

Dave Armstrong

davidaarmstrong/damisc_nodep documentation built on May 15, 2019, 6:25 p.m.