Noncentral Algorithm Assessments

• Activity Number: 616 - Multidisciplinary Advances in Computing
• Type: Contributed
• Date/Time: Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Computing
• Author(s): Jerry Lewis*
• Companies: Biogen Idec
• Keywords: accuracy; noncentral distributions; algorithms; open source software; R

Abstract

Noncentral distributions play an important role in many areas of statistics, including power and sample size analysis, tolerance intervals, Skellam distribution, normal theory distribution of CV (coefficient of variation), quantitation and detection limits of analytical methods, properties of diagnostic tests, etc. Accurate calculation of noncentral distributions can be difficult. Open source routines in R, Boost, and VBA are compared with results from proprietary routines in Maple, Mathematica, Matlab, SAS, and JMP. Specific suggestions are made for improving routines.