trend: Estimate Trend

In astsa: Applied Statistical Time Series Analysis

Estimate Trend

Description

Estimates the trend (polynomial or lowess) of a time series and returns a graphic of the series with the trend and error bounds superimposed.

Usage

trend(series, order = 1, lowess = FALSE, lowspan = .75, robust = TRUE,
         col = c(4, 6), ylab = NULL, ci=TRUE, ...)

Arguments

series	The time series to be analyzed (univariate only).
order	Order of the polynomial used to estimate the trend with a linear default (order=1) unless lowess is TRUE.
lowess	If TRUE, loess from the stats package is used to fit the trend. The default is FALSE.
lowspan	The smoother span used for lowess.
robust	If TRUE (default), the lowess fit is robust.
col	Vector of two colors for the graphic, first the color of the data (default is blue [4]) and second the color of the trend (default is magenta [6]). Both the data and trend line will be the same color if only one value is given.
ylab	Label for the vertical axis (default is the name of the series).
ci	If TRUE (default), pointwise 95
...	Other graphical parameters.

Details

Produces a graphic of the time series with the trend and a .95 pointwise confidence interval superimposed. The trend estimate and the error bounds are returned invisibly.

Value

Produces a graphic and returns the trend estimate fit and error bounds lwr and upr invisibly (see details) and with the same time series attributes as the input series.

Author(s)

D.S. Stoffer

References

You can find demonstrations of astsa capabilities at FUN WITH ASTSA.

The most recent version of the package can be found at https://github.com/nickpoison/astsa/.

In addition, the News and ChangeLog files are at https://github.com/nickpoison/astsa/blob/master/NEWS.md.

The webpages for the texts and some help on using R for time series analysis can be found at https://nickpoison.github.io/.

See Also

detrend

Examples

## Not run: 

par(mfrow=2:1)
trend(soi)
trend(soi, lowess=TRUE)

## End(Not run)

astsa documentation built on May 29, 2024, 10:29 a.m.