Abstract

In this note, we introduce a discrete counterpart of the conventional max-autoregressive moving-average process of Davis and Resnick (1989), based on the binomial thinning operator and driven by a sequence of i. i. d. nonnegative integer-valued random variables with a finite range of counts. Basic probabilistic and statistical properties of this new class of models are discussed in detail, namely the existence of a stationary distribution, and how observations’ and innovations’ distributions are related to each other. Furthermore, parameter estimation is also addressed.

Highlights

  • Modeling the temporal dependence of integer-valued time series defined on a finite range of counts, say {0, 1, . . . , n}, is nowadays a topic of research which is gaining importance in time series analysis

  • Note that traditional integer-valued ARMA-type models are useless in this context, since such models are defined over unbounded sets

  • Our aim in this paper is to construct a full integer-valued time series model with finite support, in the sense that the model should include both an autoregressive-type and a moving-average-type component ( a “full” counterpart), and it should certainly contain both a purely autoregressive-type and a purely moving-average-type model as a special case. To this extent we introduce a discrete counterpart of the conventional maxautoregressive moving-average process of Davis & Resnick (1989) which will be referred to as the maximum BARMA (in short max-BARMA(p, q)) model

Read more

Summary

Introduction

Modeling the temporal dependence of integer-valued time series defined on a finite range of counts, say {0, 1, . . . , n}, is nowadays a topic of research which is gaining importance in time series analysis. Our aim in this paper is to construct a full integer-valued time series model with finite support, in the sense that the model should include both an autoregressive-type and a moving-average-type component ( a “full” counterpart), and it should certainly contain both a purely autoregressive-type and a purely moving-average-type model as a special case. To this extent we introduce a discrete counterpart of the conventional maxautoregressive moving-average process of Davis & Resnick (1989) which will be referred to as the maximum BARMA (in short max-BARMA(p, q)) model.

The max-BARMA Model
Parameter Estimation
Conclusions
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.