cv.lo2.Rd
Function provides the cross-validation error for the loess curve in a manner that is amenable to optimization of the span.
cv.lo2( span, form, data, cost = function(y, yhat) mean((y - yhat)^2, na.rm = TRUE), K = n, numiter = 100, which = c("corrected", "raw") )
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 |
The cross-validation error from the loess curve.