BoxCox.numeric(FitAR)R Documentation

Box-Cox transformation for a time series

Description

An AR(p) model is selected using AIC and then the best Box-Cox transformation is determined. Requires package FitAR.

Usage

BoxCox.numeric(object, interval = c(-1, 1), IIDQ = TRUE, ...)

Arguments

object a vector of time series values
interval interval to be searched
IIDQ If true, IID is assumed, ie. p=0. If FALSE, AR(p) is fit with p determined using AIC.
... optional arguments

Details

If the minimum data value is <= 0, a small positive constant, equal to the negative of the minimum plus 0.25, is added to all the data values. If length(object) < 20, no AR model is used, that is, p=0.

Value

No value returned. Graphical output produced as side-effect. The plot shows relative likelihood funciton as well as the MLE and a confidence interval.

Warning

It is important not to transform the data when fitting it with arima since the optimal transformation would be found for the transformed data – not the original data. Normally this would not be a sensible thing to do.

Note

The MASS package has a similar function boxcox but this is implemented only for regression and analysis of variance.

Author(s)

A.I. McLeod

References

Box, G. E. P. and Cox, D. R. (1964) An analysis of transformations. Journal of Royal Statistical Society, Series B, vol. 26, pp. 211-246.

See Also

BoxCox.FitAR, BoxCox.Arima, BoxCox.ts

Examples

#
#annual sunspot series
BoxCox(sunspot.year, IIDQ=FALSE)
#
#non-time series example, lengths of rivers
BoxCox(rivers)


[Package FitAR version 1.0 Index]