gpr: gpr

In flowfield: Forecasts future values of a univariate time series.

Description

Interpolates the penalized regression skeleton using Gaussian process regression.

Usage

gpr(h, rec3.sd, rec3.delta, ssd, sdelta, responses)

Arguments

h	History space. Current version uses the curent and last 2 levels of the systematically determined component (SDC) and the previous 3 forward response derivatives.
rec3.sd	Most recent SDC values in the history space.
rec3.delta	Most recent forward response derivatives in the history spac.
ssd	Standard devation of the SDCs.
sdelta	Standard devation of the forward response derivatives.
responses	All values of the forward response derivatives in the history space.

Value

Returns the GPR interpolated value by comparing the current history to the entire history space.

Author(s)

Kyle A. Caudle

References

Frey, MR and Caudle, KA "Flow field forecasting for univariate time series," Statistical Analysis and Data Mining, 2013.

flowfield documentation built on May 2, 2019, 10:21 a.m.