Barcelona GSE: Graduate School of Economics

Contact Barcelona GSE|FAQ|News|Events

Home » Research » Working Papers

Barcelona GSE Working Papers Series

Barcelona GSE Research Community

Barcelona GSE Working Paper No. 669

Title: Econometrics on GPUs
Authors: Michael Creel, Sonik Mandal, Mohammad Zubair
Date: 14-11-2012
Keywords: parallel computing; graphical processing unit; GPU; econometrics; simulation-based methods; Bayesian estimation
JEL Codes: C13, C14, C15, C33
Abstract:
A graphical processing unit (GPU) is a hardware device normally used to manipulate computer memory for the display of images. GPU computing is the practice of using a GPU device for scientific or general purpose computations that are not necessarily related to the display of images. Many problems in econometrics have a structure that allows for successful use of GPU computing. We explore two examples. The first is simple: repeated evaluation of a likelihood function at different parameter values. The second is a more complicated estimator that involves simulation and nonparametric fitting. We find speedups from 1.5 up to 55.4 times, compared to computations done on a single CPU core. These speedups can be obtained with very little expense, energy consumption, and time dedicated to system maintenance, compared to equivalent performance solutions using CPUs. Code for the examples is provided.
Download this working paper in PDF format (1.13 Mb)
« List of all recent Barcelona GSE working papers
Search all working papers:
By author:
By date:
from to
By keyword(s):
By JEL code(s):
 

Home » Research » Working Papers