predict.semihregScair: Prediction of the semiparametric harmonic regression with...

In CliffordLai/harper: Harmonic Regression and Periodicity Test

Description

Return predicted values of the semiparametric harmonic regression with GAM. The prediction procedure utilizes the function predict.scair in the package scar.

Usage

## S3 method for class 'semihregScair'
predict(object, newt, ...)

Arguments

object	An object of class "semihregScair".
newt	A vector of sampled time points for predicted values.
...	Other parameters to be passed through to predicting functions.

Value

A matrix of two column, where the first column contains the sample time points and the second column contains the corresponding predicted values.

Author(s)

Yuanhao Lai

See Also

semihregScair, predict.scair.

Examples

# Simulate the series
set.seed(193)
g <- function(x){-2*exp(-x^3)}
n <- 50
t <- 1:n
lambda0 <- 0.123
phi0 <- 0.2*2*pi
u <- cos(2*pi*lambda0*t+phi0)
e <- rnorm(n)
y <- g(u)+e

# Fit the semi-harmonic regression
fit1 <- semihregScair(y,t,lambda0=lambda0,shape="in",iter=100)
names(fit1)
fit1$esty

# Prediction for the continuous time.
newt <- seq(1,n,0.05) 
head( predy <- predict(fit1,newt)) 

plot(newt,g(cos(2*pi*lambda0*newt+phi0)),
     type="b",pch=19,main="Semiparametric with scair",
     xlab="t", ylab="g(u)",
     ylim=c(min(predy[,2])-1,max(predy[,2])+0.5) )
lines(newt,predy[,2],col="red",type="b")

CliffordLai/harper documentation built on May 8, 2019, 1:53 p.m.