TSSS (version 1.2.3)

polreg: Polynomial Regression Model

Description

Estimate the trend using the AIC best polynomial regression model.

Usage

polreg(y, order, plot = TRUE, …)

Arguments

y

a univariate time series.

order

order of polynomial regression.

plot

logical. If TRUE (default), 'y' and 'trend' are plotted.

…

further arguments to be passed to plot.polreg.

Value

An object of class "polreg", which is a list with the following elements:

order.maice

MAICE (minimum AIC estimate) order.

sigma2

residual variance of the model with order \(M\). (\(0 \leq M \leq\) order \(+ 1\))

aic

AIC of the model with order \(M\). (\(0 \leq M \leq\) order \(+ 1\))

daic

AIC - minimum AIC.

coef

regression coefficients \(A(I,M)\) with order \(M\).

(\(1 \leq M \leq\) order \(+ 1\), \(1 \leq I \leq M\))

trend

trend component.

References

Kitagawa, G. (2010) Introduction to Time Series Modeling. Chapman & Hall/CRC.

Examples

# NOT RUN {
# The daily maximum temperatures for Tokyo
data(Temperature)
polreg(Temperature, order = 7)

# Wholesale hardware data
data(WHARD)
y <- log10(WHARD)
polreg(y, order = 15)
# }