Exponential Smoothing for Time Series with Outliers

Abstract

A simple robustification of Kalman filter is suggested using a simple truncation of the recursive residuals. This concept is applied mainly to various types of exponential smoothing but recursive estimation in Box-Jenkins methodology is also mentioned.

The methods are demonstrated using simulated data.

Keywords

• Exponential smoothing
• Kalman filter
• Outliers
• Robust smoothing and forecasting