Parallelization Experience with Four Canonical Econometric Models Using ParMitISEM
This paper presents the parallel computing implementation of the MitISEM algorithm, labeled Parallel MitISEM. The basic MitISEM algorithm provides an automatic and flexible method to approximate a non-elliptical target density using adaptive mixtures of Student-t densities, where only a kernel of th...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2016-03-01
|
| Series: | Econometrics |
| Subjects: | |
| Online Access: | http://www.mdpi.com/2225-1146/4/1/11 |
| _version_ | 1828150055992819712 |
|---|---|
| author | Nalan Baştürk Stefano Grassi Lennart Hoogerheide Herman K. van Dijk |
| author_facet | Nalan Baştürk Stefano Grassi Lennart Hoogerheide Herman K. van Dijk |
| author_sort | Nalan Baştürk |
| collection | DOAJ |
| description | This paper presents the parallel computing implementation of the MitISEM algorithm, labeled Parallel MitISEM. The basic MitISEM algorithm provides an automatic and flexible method to approximate a non-elliptical target density using adaptive mixtures of Student-t densities, where only a kernel of the target density is required. The approximation can be used as a candidate density in Importance Sampling or Metropolis Hastings methods for Bayesian inference on model parameters and probabilities. We present and discuss four canonical econometric models using a Graphics Processing Unit and a multi-core Central Processing Unit version of the MitISEM algorithm. The results show that the parallelization of the MitISEM algorithm on Graphics Processing Units and multi-core Central Processing Units is straightforward and fast to program using MATLAB. Moreover the speed performance of the Graphics Processing Unit version is much higher than the Central Processing Unit one. |
| first_indexed | 2024-04-11T21:37:51Z |
| format | Article |
| id | doaj.art-ba22d8be6eb14cfca5974afd046e0359 |
| institution | Directory Open Access Journal |
| issn | 2225-1146 |
| language | English |
| last_indexed | 2024-04-11T21:37:51Z |
| publishDate | 2016-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Econometrics |
| spelling | doaj.art-ba22d8be6eb14cfca5974afd046e03592022-12-22T04:01:41ZengMDPI AGEconometrics2225-11462016-03-01411110.3390/econometrics4010011econometrics4010011Parallelization Experience with Four Canonical Econometric Models Using ParMitISEMNalan Baştürk0Stefano Grassi1Lennart Hoogerheide2Herman K. van Dijk3Department of Quantitative Economics, School of Business and Economics, Maastricht University, Maastricht 6211LM, The NetherlandsSchool of Economics, Keynes College, University of Kent, Canterbury CT27NP, UKDepartment of Econometrics and Tinbergen Institute, Vrije Universiteit Amsterdam, Amsterdam 1081HV, The NetherlandsDepartment of Econometrics and Tinbergen Institute, Vrije Universiteit Amsterdam, Amsterdam 1081HV, The NetherlandsThis paper presents the parallel computing implementation of the MitISEM algorithm, labeled Parallel MitISEM. The basic MitISEM algorithm provides an automatic and flexible method to approximate a non-elliptical target density using adaptive mixtures of Student-t densities, where only a kernel of the target density is required. The approximation can be used as a candidate density in Importance Sampling or Metropolis Hastings methods for Bayesian inference on model parameters and probabilities. We present and discuss four canonical econometric models using a Graphics Processing Unit and a multi-core Central Processing Unit version of the MitISEM algorithm. The results show that the parallelization of the MitISEM algorithm on Graphics Processing Units and multi-core Central Processing Units is straightforward and fast to program using MATLAB. Moreover the speed performance of the Graphics Processing Unit version is much higher than the Central Processing Unit one.http://www.mdpi.com/2225-1146/4/1/11Importance samplingparallel computingMitISEMMCMC |
| spellingShingle | Nalan Baştürk Stefano Grassi Lennart Hoogerheide Herman K. van Dijk Parallelization Experience with Four Canonical Econometric Models Using ParMitISEM Econometrics Importance sampling parallel computing MitISEM MCMC |
| title | Parallelization Experience with Four Canonical Econometric Models Using ParMitISEM |
| title_full | Parallelization Experience with Four Canonical Econometric Models Using ParMitISEM |
| title_fullStr | Parallelization Experience with Four Canonical Econometric Models Using ParMitISEM |
| title_full_unstemmed | Parallelization Experience with Four Canonical Econometric Models Using ParMitISEM |
| title_short | Parallelization Experience with Four Canonical Econometric Models Using ParMitISEM |
| title_sort | parallelization experience with four canonical econometric models using parmitisem |
| topic | Importance sampling parallel computing MitISEM MCMC |
| url | http://www.mdpi.com/2225-1146/4/1/11 |
| work_keys_str_mv | AT nalanbasturk parallelizationexperiencewithfourcanonicaleconometricmodelsusingparmitisem AT stefanograssi parallelizationexperiencewithfourcanonicaleconometricmodelsusingparmitisem AT lennarthoogerheide parallelizationexperiencewithfourcanonicaleconometricmodelsusingparmitisem AT hermankvandijk parallelizationexperiencewithfourcanonicaleconometricmodelsusingparmitisem |