Computing and Graphics in Statisticsby ANDREAS BUJA
Pub. Date: 01/28/1997
Publisher: Springer New York
Computing and Graphics in Statistics presents issues that arise in the development of integrated statistical software systems which have led to the adaptation of ideas from computer science, particularly programming environments, programming paradigms, and artificial intelligence. Examples are given that distinguish statistics from many physical sciences such as genuine high-dimensional objects - multivariate data or functions of many variables. Demonstrates automatic methods for finding reasonable domains and ranges for plotting univariate functions. Deals with computer intensive methodology including the bootstrap method, Bayesian inferrence and its associated integration problems.
Table of ContentsAn interface between S and Mathematica.- Looking at large data sets using binned data plots.- Integrating a robust option into a multiple regression computing environment.- Algorithm development for nonstandard least squares problems. Repeated categorical responses with missing values: A case study.- Importance sampling for Bayesian estimation.- A software model for statistical graphics.- Geometric abstractions for constrained optimization of layouts.- Construction of line densities for parallel coordinate plots.- GLIMPSE, a knowledge-based front end for GLIM.- GENSTAT as a computing environment.- Situations, summaries and model objects.- On estimation and visualization of higher dimensional surfaces.- Odds plots: A graphical aid for finding associations between views of a data set.- A Shastic Approach to Load Balancing in Coarse Grain Parallel Computers.- Algorithms for choosing the domain and range when plotting a function.- High-dimensional depth-cuing for guided tours of multivariate data.- Towards a structured data analysis environment: A cognition-based design.
and post it to your social network
Most Helpful Customer Reviews
See all customer reviews >