Outliers in multivariate time series

By Ruey S. Tsay, D. Pena, Alan E. Pankratz. 2000

189 citations recorded by the source corpus.

This paper considers outliers in multivariate time series analysis. It generalizes four types of disturbances commonly used in the univariate time series analysis to the multivariate case, and investigates dynamic effects of a multivariate outlier on individual components if marginal models are used. An innovational outlier of a vector series can introduce a patch of outliers for the marginal component models. The paper also proposes an iterative procedure to detect and handle multiple outliers. By comparing and contrasting results of univariate and multivariate outlier detections, one can gain insights into the characteristics of an outlier. An outlier in a component series mayor may not have significant impacts on the other components. We use real examples to demonstrate the proposed analysis.

Subfields: Finance, Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/9a7e0641b0d47145e1ea3742b6bafca7a786128e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Finance

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Anomaly detection: A survey (2009, methodology): “Time-series data has been extensively explored in the contextual anomaly detection category [Abraham and Chuang 1989; Abraham and Box 1979; Rousseeuw and Leroy 1987; Bianco et al. 2001; Fox 1972; Salvador and Chan 2003; Tsay et al. 2000; Galeano et al. 2004; Zeevi et al. 1997].”
Robust Forecasting of Dynamic Conditional Correlation GARCH Models (2013, background): “As noted e.g. in Tsay, Peña, and Pankratz (2000), there exist other definitions of outliers in multivariate time series.”
Outlier Detection for Temporal Data: A Survey (2014, background): “[78] propose a vector ARIMA model to identify additive outliers, innovation outliers, level shifts, and temporary changes from multi-variate time series.”
Covariance changes detection in multivariate time series (2007, methodology): “Third, in both algorithms we require a minimum distance between changes, larger than d. If several changes are found in an interval smaller than d, these changes will be considered as outliers and estimated by the procedure proposed in Tsay et al. (2000) .”
Cointegration Analysis in the Presence of Outliers (2004, unknown): “To determine the location as well as the type of outliers, an automatic outlier detection procedure can be used (see also Tsay 1986; Chen and Liu 1993; Tsay et al. 2000).”
The Identification of Multiple Outliers in ARIMA Models (2003, background): “Tsay et al. (2000) have shown that single multivariate outliers can produce patches of outliers in the univariate time series.”