Search

Examines the accuracy of sample moments calculation using statistical programs. Advantages of the desk calculator algorithm; Use of desk calculator algorithm in short word length machines; Differences in the accuracy of algorithms for computing sample moments.

Presents several developments in statistical computing in the United States as of November 1989. Development of computer programs for limiting and serial dilution assays; Introduction of Kwiskat (version 2.00); Release of Version B-02 of the MADMANager Utility Programs.

Presents statistical computer programs in the United States. Availability of PCBRAP down-loaded version of the Bayesian Analysis Package; Methodologies used in the MUDIFT program for the distribution-free comparison of growth curves; Calculation of percentiles through PCT.

Reports developments in the field of statistical computing as of August 1988. Updates on computer software developments; Presentation of computing products to serve as aids to statisticians; Summaries of selected committee reports dealing with statistical computing.

Focuses on the issues on statistical community. Error in regression diagnostics using residual plots; Critical values of the lilliefors statistic; Basis for the principle of minimum discrimination information.

Presents letters to the editor in the August 1987 issue of The American Statistician journal. Criticism on the use of high technology matrix wrench; Failure on the recognition of references on the 1982 article of The American Statistician Journal; Apology for the misstated article.

In Francis Galton's 1886 article "Regression Towards Mediocrity in Hereditary Stature," Galton analyzed the heights of 928 adult children and their 205 pairs of parents to illustrate his linear regression model. Although Galton's data have been recalled frequently to illustrate linear regression and regression toward the mean, no one seems to have noticed that his height data do not fit his model. The purpose of this article is both to reveal this curiosity and to find a possible explanation for its existence using related data from Galton's colleague Karl Pearson. [ABSTRACT FROM AUTHOR]

Presents an analytic approximation to the distribution of test statistic for normality of H.W. Lilliefors. Application of Lilliefors of Kolmogorov-Smirnov statistic for testing normality; Occurrence of discrepancies; Decrease in the critical values of test statistic as the inverse of sample size square root.

Presents letters to the editor regarding statistical studies. Description of problems with single-precision algorithms; Presentation of errors in widely used statistical packages; Comments on the fundamentals of practical statistics.

Comments on the need for instructors to take a significant fraction of time from statistics courses to teach the use of statistical computer packages. Situations cited by Searle when caution is needed; Opportunity for statisticians to educate others in the proper use of statistical techniques.

Focuses on least median of squares regression. Use of this method to minimize the median squared difference between observed and fitted response; Regression surface in the closest pair of parallel hyperplanes that contain 50 percent of the data.

Presents the STATOOLS, a statistical utility programs. Use of studentized range statistics; Functions of the programs; Features of the programs.

Features the computer program PC-SIZE. Functions of the program; Number of source code in PC-SIZE; Price of the product; Use of the program for obtaining the common sample size required to test the equality of two proportions.