ThetaFn: Optimize theta

In nwfscNLTS: Non-linear time series

Description

Estimate optimal degree of non-linearity for S-map

Usage

ThetaFn(Y, PredInterval, ThetaSet = c(0, exp(seq(-3, 3))), Nembed)

Arguments

Y	Time series
PredInterval	Prediction interval, i.e. 1 represents predicting one observation ahead
ThetaSet	Vector containing candidate set of theta's to seach across
Nembed	Embedding dimension

Value

Returns a list with the crossvalidation correlation for each candidate theta, as well as the theta that maximizes that correlation

Author(s)

James Thorson

References

Sugihara, G., and May, R.M. 1990. Nonlinear forecasting as a way of distinguishing chaos from measurement error in time series. Nature 334: 734-741. Sugihara, G. 1994. Nonlinear forecasting for the classification of natural time series. Philosophical Transactions of the Royal Society B: Biological Sciences 348: 477-495.

nwfscNLTS documentation built on May 2, 2019, 4:31 p.m.