Influential Observations in Time Series.
Abstract
This paper studies how to identify influential observations in univariate ARIMA time series models and how to measure their effects on the estimated parameters of the model. The sensitivity of the parameters to the presence of either additive or innovational outliers is analyzed, and influence statistics based on the Mahalanobis distance are presented. The statistic linked to additive outliers is shown to be very useful to indicate the robustness of the fitted model to the given data set. Its application is illustrated using simulation results and a relevant set of historical data. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1984
- Accession Number
- ADA144719
Entities
People
- D. Pena
Organizations
- University of Wisconsin–Madison