
Accession Number : ADA132722
Title : A User's Guide to BISAM (BIvariate SAMple): The Bivariate Data Modeling Program.
Descriptive Note : Technical rept.,
Corporate Author : TEXAS A AND M UNIV COLLEGE STATION DEPT OF STATISTICS
Personal Author(s) : Woodfield,Terry J
PDF Url : ADA132722
Report Date : Aug 1983
Pagination or Media Count : 68
Abstract : The BISAM (BIvariate SAMple) program is a FORTRAN program composed of a main program and 37 subprograms. Several of the subprograms come from ONESAM and the TIMESBOARD Time Series Subroutine Library. The main feature of BISAM is the ability to provide estimates of the bivariate densityquantile function of a set of bivariate data. A univariate analysis is also provided similar to that of ONESAM but with considerably less output. The univariate analysis is not intended to replace a ONESAM analysis if such a detailed analysis is required. The bivariate analysis includes display of various nonparametric measures of association along with some entropy measures that provide diagnostics for model selection and testing for independence. Graphical display of the estimated bivariate dependence density and densityquantile is the responsibility of the user, but a two step procedure utilizing the GCONTOUR and G3D procedures of SAS/GRAPH will be suggested in a later section. Different installations will have different software to provide three dimensional plots, hence the plotting option has been excluded from BISAM. One of the advantages of the modeling approach adopted by the BISAM program is the ability to obtain quickly and efficiently a set of function values for a dense grid of bivariate points promoting the examination of three dimensional representation of functions of interest. (Author)
Descriptors : *Computer programs, *Bivariate analysis, *User manuals, Estimates, Fortran, Algorithms, Nonparametric statistics, Fourier analysis, Plotting, Entropy, Density, Exponential functions, Mathematical models, Three dimensional
Subject Categories : Statistics and Probability
Computer Programming and Software
Distribution Statement : APPROVED FOR PUBLIC RELEASE