Faculty Advisor

Petruccelli, Joseph D.

Abstract

The accuracy and stability of the Ti-83+ graphing calculators and the International Mathematical and Statistical Libraries were assessed. Random number generators were tested using Marsaglia's DIEHARD Battery of Randomness Tests; NIST Statistical Reference Datasets were used to assess the accuracy of univariate summary statistics, analysis of variance, linear regression and nonlinear regression; probability distribution functions were assessed with Knusel's ELV. The results of the probablity distribution tests show that all of the statistical packages' distribution functions compute inaccurate probabilities coresponding to extreme lower quantiles. None of the statistical packages passed all of the DIEHARD randomness tests. All of the packages computed accurate univariate summary statistics, and mostly accurate analysis of variance. Nonlinear regression tests showed that ISML's nonlinear regression routine becomes unstable when the regression solutions are in the vicinity of zero.

Publisher

Worcester Polytechnic Institute

Date Accepted

January 2004

Major

Mathematical Sciences

Project Type

Major Qualifying Project

Accessibility

Restricted-WPI community only

Advisor Department

Mathematical Sciences

Share

COinS