Using GPU Simulation to Accurately Fit to the Power-Law Distribution
Computation
2013-05-30 v1 Distributed, Parallel, and Cluster Computing
Computational Physics
Data Analysis, Statistics and Probability
Applications
Abstract
This article describes a methodology for fitting experimental data to the discrete power-law distribution and provides the results of a detailed simulation exercise used to calculate accurate cutoff values used to assess the fit to a power-law distribution when using the maximum likelihood estimation for the exponent of the distribution. Using massively parallel programming computing, we were able to accelerate by a factor of 60 the computational time required for these calculations across a range of parameters and construct a series of detailed tables containing the test values to be used in a Kolmogorov-Smirnov goodness-of-fit test, allowing for an accurate assessment of the power-law fit from empirical data.
Cite
@article{arxiv.1305.6738,
title = {Using GPU Simulation to Accurately Fit to the Power-Law Distribution},
author = {Efstratios Rappos and Stephan Robert},
journal= {arXiv preprint arXiv:1305.6738},
year = {2013}
}