animate,deseats-method {deseats}

Animate Locally Weighted Regression Results

Description

The results of locally weighted regression results acquired through decomposition of seasonal time series via the function deseats can be animated automatically.

Usage

## S4 method for signature 'deseats'
animate(
  object,
  col.obs = "grey74",
  col.fit = "red",
  col.weights = "#00D40E",
  col.window = "deepskyblue4",
  col.spot = "orange",
  save = NULL,
  xlab = "Time",
  ylab1 = "Estimated trend + seasonality",
  ylab2 = "Active kernel weights",
  main = NULL,
  ...
)

Arguments

object

an object of class "deseats".

col.obs

the color to use for the observations.

col.fit

the color to use for the fitted values.

col.weights

the color to use for the active kernel weights.

col.window

the color to use for the window defined through the bandwidth

col.spot

the color to highlight the estimation time point.

save

whether to save the animation or not; for NULL, it will not be saved; to save it in the current working directory, use either save = "PDF" or save = "HTML", which also specifies the file format.

xlab

the label for the x-axis.

ylab1

the label for the y-axis on the left-hand side.

ylab2

the label for the second y-axis on the right-hand side.

main

the plot title.

...

currently without use; implemented for possible future compatibility.

Details

deseats estimation results are automatically animated through this method. It shows the observed series together with fitted values (trend + seasonality), the smoothing window, the fitted values from the local regression, and the active kernel weights.

Value

The function returns NULL.

Author(s)

• Dominik Schulz (Research Assistant) (Department of Economics, Paderborn University),
  Author and Package Creator

Examples

### Creating the animation might take a while
Xt <- log(EXPENDITURES)
smoothing_options <- set_options(order_poly = 3)
est <- deseats(Xt, smoothing_options = smoothing_options)
animate(est)