Courses in statistics and experimental design can be enhanced through use of crafted data sets. The use of examples highlights the interface between data and statistical routine. FORTRAN programs utilizing the International Mathematical and Statistical Library subroutines permit the user to control the variance—covariance structure of multivariate normal variables and build data sets that have instructional value. Scale transformations and Monte Carlo simulations of the data can be performed as well.
Get full access to this article
View all access options for this article.
References
1.
HealyM. J. (1968). Algorithm AS 6: Triangular decomposition of a symmetric matrix. Applied Statistics, 17, 195–197.
2.
KennedyW. J.Jr.GentileJ. E. (1980). Statistical computing. New York, NY: Marcel Dekker.
3.
KnuthD. (1983). The art of computing: Semi-numerical algorithms. Reading, MA: Addison-Wesley.
4.
MaindonaldJ. H. (1984). Statistical computation. New York: Wiley.
5.
SearleS. S. (1989). Statistical computing packages: Some words of caution. American Statistician, 43, 189–190.
6.
WilkinsonL. (1990). SYSTAT: A system for statistics. Evanston, IL: SYSTAT, Inc.
