Activity Number:
|
338
- Time Series and Forecasting
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Business and Economic Statistics Section
|
Abstract #323228
|
View Presentation
|
Title:
|
The Gini Autocovariance Function Applied to Heavy Tailed Linear Time Series
|
Author(s):
|
Marcel Carcea* and Robert Serfling
|
Companies:
|
Western New England University and University of Texas at Dallas
|
Keywords:
|
linear time series ;
autocovariance ;
heavy tails
|
Abstract:
|
Heavy tailed distributions and data are becoming of greater importance in Economics, Finance and Actuarial Science. Hence developing new techniques and/or adapting old techniques to studying heavy tailed distributions is more than just a statistical curiosity. Here we present a "Gini autocovariance" in the context of linear time series models. The main advantage of "Gini autocovariance" approach is that it operates under only first moment assumptions. Also "Gini autocovariance" has population analog, whereas the population Pearson autocovariance does not exist. This talk presents results on the performance of "Gini autocovariance" via simulation studies allowing a wide range of typical innovation and outlier scenarios in the context of the linear time series models. It is seen that the "Gini" approach competes well with standard methods and provides a new reliable tool in time series modeling in heavy tailed settings.
|
Authors who are presenting talks have a * after their name.