Activity Number:
|
492
- Recent Advances in Modeling Complex Dependent Data
|
Type:
|
Invited
|
Date/Time:
|
Wednesday, August 1, 2018 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Business and Economic Statistics Section
|
Abstract #326716
|
Presentation
|
Title:
|
Count Time Series Models Based on Expectation Thinning Operators
|
Author(s):
|
Harry Joe*
|
Companies:
|
University of British Columbia
|
Keywords:
|
count time series;
thinning operator
|
Abstract:
|
Expectation thinning operators for count time series are compounding operators that include the binomial thinning operator as a special case. Some new results are given for the relationship of the stationary marginal distribution, the innovation distribution and the family of compounding operators that satisfy a self-generalized closure property. It will be shown how to build integer-autoregressive time series models that can accommodate covariates. A data example will be presented as well as the numerical methods required for parameter estimation.
|
Authors who are presenting talks have a * after their name.