Computation of the Integral of the Bivariate Normal Distribution Over Arbitrary Polygons

Abstract

An efficient automatic procedure is given for evaluating the integral of the bivariate normal density function (IBND) over an arbitrary polygon Pi. The polygon Pi, defined by N points, falls into one or more of the following classes: <S>, simple polygons; <S-bar>, limit elements of sequences of uniformly bounded N-sided simple polygons of the same orientation; <Pi>, arbitrary polygons, including self-intersecting (SI) ones, where <S> coincides with or belongs to <S-bar> which coincides with or belongs to <Pi>. It is not necessary to specify the class beforehand. The method extracts from Pi a set of N exterior angular regions. The IBND is evaluated over each of these, and the results are properly combined to yield IBND for Pi. In case Pi is SI, account must be taken of the number of its 'primary circuits' and their orientations. A by-product of the analyses is the evaluation of a function A(Pi). Another procedure for obtaining the same final results is described for completeness which is not as efficient. It treats an SI polygon by decomposing it into a finite set of S or S-bar type elements. The IBND is evaluated over each of these; the results are properly summed to give the IBND for Pi. In contrast to the first method, the smallest class <S>, <S-bar>, <Pi> to which Pi belongs must be specified for computational efficiency. The Fortran IV programs for both procedures are presently set to yield approximately 3, 6, or 9-decimal-digit accuracy. Fortran IV listings of the programs are given.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1980
Accession Number
ADA102466

Entities

People

  • A. R. Didonato
  • R. K. Hageman

Organizations

  • Naval Surface Warfare Center Dahlgren Division

Tags

Communities of Interest

  • Air Platforms
  • C4I
  • Materials and Manufacturing Processes
  • Weapons Technologies

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Computer Programs
  • Computer Science
  • Computers
  • Data Science
  • Information Science
  • Integrals
  • Navy
  • Normal Density Functions
  • Operations Research
  • Polygons
  • Probability
  • Probability Distributions
  • Random Variables
  • Sequences
  • Statistics

Readers

  • Graph Algorithms and Convex Optimization.
  • Regression Analysis.