ts_fil_winsor: Winsorization of Time Series

In tspredit: Time Series Prediction with Integrated Tuning

Winsorization of Time Series

Description

Apply Winsorization to limit extreme values by replacing them with nearer order statistics, reducing the influence of outliers.

Usage

ts_fil_winsor()

Value

A ts_fil_winsor object.

References

• J. W. Tukey (1962). The future of data analysis. Annals of Mathematical Statistics. (Winsorization discussed in robust summaries.)

Examples

# Winsorization: cap extreme values to reduce outlier impact
# Load package and example data
library(daltoolbox)
data(tsd)
tsd$y[9] <- 2 * tsd$y[9]  # inject an outlier

# Fit Winsor filter and transform series
filter <- ts_fil_winsor()
filter <- fit(filter, tsd$y)
y <- transform(filter, tsd$y)

# Plot original vs Winsorized series
plot_ts_pred(y = tsd$y, yadj = y)

tspredit documentation built on Feb. 11, 2026, 9:08 a.m.