scientific article; zbMATH DE number 1335868
Statistical tables have been important tools for use by research workers throughout the history of statistics. Changes in computing environment have occurred in the past and are likely to continue in the future to change the ways in which tables are presented, but the basic need for accurate approximations to various function values will persist. Accuracy in approximation, especially accuracy to a guaranteed stated level, has always been and continues to be very difficult to achieve in most important tabling activities. Frequently, even today, tables are published wherein not all entries can be guaranteed to be correctly rounded approximations to the true value for the number of significant digits shown in the table. Regardless of how tables are presented, this possible deficiency in accuracy of tabular product should not be allowed in today's computing environment. Methodology exists that will assist in this regard, but the statistical computing community has been slow to recognize this fact and to take advantage of it when preparing statistical tables. The purpose of this chapter is to describe, and demonstrate the use of potentially useful numerical tools that we believe should no longer be ignored in statistical computing. The entire collection of numerical tools is usually referred to as interval analysis. For the entire collection see [Zbl 0913.00046].
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4258557)
