Computationally Efficient Poisson Time-Varying Autoregressive Models through Bayesian Lattice Filters
Estimation of time-varying autoregressive models for count-valued time series can be computationally challenging. In this direction, we propose a time-varying Poisson autoregressive (TV-Pois-AR) model that accounts for the changing intensity of the Poisson process. Our approach can capture the laten...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-10-01
|
| Series: | Stats |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2571-905X/6/4/65 |
| _version_ | 1797379281665392640 |
|---|---|
| author | Yuelei Sui Scott H. Holan Wen-Hsi Yang |
| author_facet | Yuelei Sui Scott H. Holan Wen-Hsi Yang |
| author_sort | Yuelei Sui |
| collection | DOAJ |
| description | Estimation of time-varying autoregressive models for count-valued time series can be computationally challenging. In this direction, we propose a time-varying Poisson autoregressive (TV-Pois-AR) model that accounts for the changing intensity of the Poisson process. Our approach can capture the latent dynamics of the time series and therefore make superior forecasts. To speed up the estimation of the TV-AR process, our approach uses the Bayesian Lattice Filter. In addition, the No-U-Turn Sampler (NUTS) is used, instead of a random walk Metropolis–Hastings algorithm, to sample intensity-related parameters without a closed-form full conditional distribution. The effectiveness of our approach is evaluated through model-based and empirical simulation studies. Finally, we demonstrate the utility of the proposed model through an example of COVID-19 spread in New York State and an example of US COVID-19 hospitalization data. |
| first_indexed | 2024-03-08T20:21:54Z |
| format | Article |
| id | doaj.art-480a03a69f654acbaf296dedac016650 |
| institution | Directory Open Access Journal |
| issn | 2571-905X |
| language | English |
| last_indexed | 2024-03-08T20:21:54Z |
| publishDate | 2023-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Stats |
| spelling | doaj.art-480a03a69f654acbaf296dedac0166502023-12-22T14:43:16ZengMDPI AGStats2571-905X2023-10-01641037105210.3390/stats6040065Computationally Efficient Poisson Time-Varying Autoregressive Models through Bayesian Lattice FiltersYuelei Sui0Scott H. Holan1Wen-Hsi Yang2Amazon.com, Inc., New York, NY 10001, USADepartment of Statistics, University of Missouri, Columbia, MO 65211-6100, USASchool of Agriculture and Food Sciences, The University of Queensland, St Lucia, QLD 4072, AustraliaEstimation of time-varying autoregressive models for count-valued time series can be computationally challenging. In this direction, we propose a time-varying Poisson autoregressive (TV-Pois-AR) model that accounts for the changing intensity of the Poisson process. Our approach can capture the latent dynamics of the time series and therefore make superior forecasts. To speed up the estimation of the TV-AR process, our approach uses the Bayesian Lattice Filter. In addition, the No-U-Turn Sampler (NUTS) is used, instead of a random walk Metropolis–Hastings algorithm, to sample intensity-related parameters without a closed-form full conditional distribution. The effectiveness of our approach is evaluated through model-based and empirical simulation studies. Finally, we demonstrate the utility of the proposed model through an example of COVID-19 spread in New York State and an example of US COVID-19 hospitalization data.https://www.mdpi.com/2571-905X/6/4/65Bayesian hierarchical modelnonstationary time seriespartial autocorrelationtime-varying spectral densityvector autoregressive model |
| spellingShingle | Yuelei Sui Scott H. Holan Wen-Hsi Yang Computationally Efficient Poisson Time-Varying Autoregressive Models through Bayesian Lattice Filters Stats Bayesian hierarchical model nonstationary time series partial autocorrelation time-varying spectral density vector autoregressive model |
| title | Computationally Efficient Poisson Time-Varying Autoregressive Models through Bayesian Lattice Filters |
| title_full | Computationally Efficient Poisson Time-Varying Autoregressive Models through Bayesian Lattice Filters |
| title_fullStr | Computationally Efficient Poisson Time-Varying Autoregressive Models through Bayesian Lattice Filters |
| title_full_unstemmed | Computationally Efficient Poisson Time-Varying Autoregressive Models through Bayesian Lattice Filters |
| title_short | Computationally Efficient Poisson Time-Varying Autoregressive Models through Bayesian Lattice Filters |
| title_sort | computationally efficient poisson time varying autoregressive models through bayesian lattice filters |
| topic | Bayesian hierarchical model nonstationary time series partial autocorrelation time-varying spectral density vector autoregressive model |
| url | https://www.mdpi.com/2571-905X/6/4/65 |
| work_keys_str_mv | AT yueleisui computationallyefficientpoissontimevaryingautoregressivemodelsthroughbayesianlatticefilters AT scotthholan computationallyefficientpoissontimevaryingautoregressivemodelsthroughbayesianlatticefilters AT wenhsiyang computationallyefficientpoissontimevaryingautoregressivemodelsthroughbayesianlatticefilters |