Date of Award

5-2015

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Economics and Finance

Committee Chair(s)

Tyler Brough

Committee

Tyler Brough

Committee

Jason Smith

Committee

Jared DeLisle

Abstract

This Thesis explores the use of different programming paradigms, platforms and languages to maximize the speed of convergence in Financial Derivatives Models. The study focuses on the strengths and drawbacks of various libraries and their associated languages as well as the difficulty of importing code into massively parallel processes. Key languages utilized in this project are C++, Python, Java, C, and Matlab. Along with these languages, several multiprocessing libraries such as C++ AMP, Python NumbaPro, Numpy, Anaconda, Finance Toolkit and such were compared. The results of the development and implementation of code suggests that once we massively parallelize the operations, the only difference in speeds across systems are found only in the unique language's overhead. The findings also suggest that in terms of development and testing time, one should choose to utilize a language they are most familiar with rather than to try a minimalistic approach using a more base level language.

Included in

Business Commons

Share

COinS