Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB Toolbox
This paper presents the MATLAB package DeCo (density combination) which is based on the paper by Billio, Casarin, Ravazzolo, and van Dijk (2013) where a constructive Bayesian approach is presented for combining predictive densities originating from different models or other sources of information. T...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Foundation for Open Access Statistics
2015-11-01
|
| Series: | Journal of Statistical Software |
| Subjects: | |
| Online Access: | https://www.jstatsoft.org/index.php/jss/article/view/1135 |
| _version_ | 1818207405258833920 |
|---|---|
| author | Roberto Casarin Stefano Grassi Francesco Ravazzolo Herman K. van Dijk |
| author_facet | Roberto Casarin Stefano Grassi Francesco Ravazzolo Herman K. van Dijk |
| author_sort | Roberto Casarin |
| collection | DOAJ |
| description | This paper presents the MATLAB package DeCo (density combination) which is based on the paper by Billio, Casarin, Ravazzolo, and van Dijk (2013) where a constructive Bayesian approach is presented for combining predictive densities originating from different models or other sources of information. The combination weights are time-varying and may depend on past predictive forecasting performances and other learning mechanisms. The core algorithm is the function DeCo which applies banks of parallel sequential Monte Carlo algorithms to filter the time-varying combination weights. The DeCo procedure has been implemented both for standard CPU computing and for graphical process unit (GPU) parallel computing. For the GPU implementation we use the MATLAB parallel computing toolbox and show how to use general purpose GPU computing almost effortlessly. This GPU implementation provides a speed-up of the execution time of up to seventy times on a standard CPU MATLAB implementation on a multicore CPU. We show the use of the package and the computational gain of the GPU version through some simulation experiments and empirical applications. |
| first_indexed | 2024-12-12T04:28:24Z |
| format | Article |
| id | doaj.art-c497288b5456402fb1e7f948a20bb94d |
| institution | Directory Open Access Journal |
| issn | 1548-7660 |
| language | English |
| last_indexed | 2024-12-12T04:28:24Z |
| publishDate | 2015-11-01 |
| publisher | Foundation for Open Access Statistics |
| record_format | Article |
| series | Journal of Statistical Software |
| spelling | doaj.art-c497288b5456402fb1e7f948a20bb94d2022-12-22T00:38:08ZengFoundation for Open Access StatisticsJournal of Statistical Software1548-76602015-11-0168113010.18637/jss.v068.i03961Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB ToolboxRoberto CasarinStefano GrassiFrancesco RavazzoloHerman K. van DijkThis paper presents the MATLAB package DeCo (density combination) which is based on the paper by Billio, Casarin, Ravazzolo, and van Dijk (2013) where a constructive Bayesian approach is presented for combining predictive densities originating from different models or other sources of information. The combination weights are time-varying and may depend on past predictive forecasting performances and other learning mechanisms. The core algorithm is the function DeCo which applies banks of parallel sequential Monte Carlo algorithms to filter the time-varying combination weights. The DeCo procedure has been implemented both for standard CPU computing and for graphical process unit (GPU) parallel computing. For the GPU implementation we use the MATLAB parallel computing toolbox and show how to use general purpose GPU computing almost effortlessly. This GPU implementation provides a speed-up of the execution time of up to seventy times on a standard CPU MATLAB implementation on a multicore CPU. We show the use of the package and the computational gain of the GPU version through some simulation experiments and empirical applications.https://www.jstatsoft.org/index.php/jss/article/view/1135density forecast combinationsequential Monte Carloparallel computingGPUMATLAB |
| spellingShingle | Roberto Casarin Stefano Grassi Francesco Ravazzolo Herman K. van Dijk Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB Toolbox Journal of Statistical Software density forecast combination sequential Monte Carlo parallel computing GPU MATLAB |
| title | Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB Toolbox |
| title_full | Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB Toolbox |
| title_fullStr | Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB Toolbox |
| title_full_unstemmed | Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB Toolbox |
| title_short | Parallel Sequential Monte Carlo for Efficient Density Combination: The DeCo MATLAB Toolbox |
| title_sort | parallel sequential monte carlo for efficient density combination the deco matlab toolbox |
| topic | density forecast combination sequential Monte Carlo parallel computing GPU MATLAB |
| url | https://www.jstatsoft.org/index.php/jss/article/view/1135 |
| work_keys_str_mv | AT robertocasarin parallelsequentialmontecarloforefficientdensitycombinationthedecomatlabtoolbox AT stefanograssi parallelsequentialmontecarloforefficientdensitycombinationthedecomatlabtoolbox AT francescoravazzolo parallelsequentialmontecarloforefficientdensitycombinationthedecomatlabtoolbox AT hermankvandijk parallelsequentialmontecarloforefficientdensitycombinationthedecomatlabtoolbox |