A Practical Guide to Parallelization in Economics

Abstract

The pdf file of the paper includes several embedded links of interest. The Github repository of the code used in the paper is: https://github.com/davidzarruk/Parallel_Computing. We thank many cohorts of students at the University of Pennsylvania who have endured earlier drafts of the slides behind this guide and pointed out ways to improve them. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. At least one co-author has disclosed a financial relationship of potential relevance for this research. Further information is available online at http://www.nber.org/papers/w24561.ack NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

A Framework for Solving Non-Linear DSGE Models

“…One further step would involve constructing an easy to use toolbox, including different options for basis functions, kernels and general options within our framework. Moreover, our framework also benefits from the recent tools of parallelization (see Fernández-Villaverde and Valencia (2018)), which opens the possibility of optimizing the algorithms.…”

Econometrics at Scale: Spark up Big Data in Economics

“…The major advantage of parallel computing is that it can allow to solve quantitative problems, that were previously prohibitive expensive to evaluate. Fernández-Villaverde and Valencia (2018) provide an excellent overview of parallel computing performance of various programming languages and illustrate the runtime gains for solving a standard value function iteration problem. They point out that (depending on your problem) you can speed up the analysis by a factor equal to the number of your computer's CPU cores.…”

Estimating DSGE Models: Recent Advances and Future Challenges

“…Fernández-Villaverde et al ( 2016) is an updated survey of the main existing solution methods for DSGE models. Fernández-Villaverde and Valencia (2018) outline how to parallelize these methods. 1 All of these conditional densities can be exploited to take the model to the data.…”