On Generating Inverse Gaussian Variates.
Abstract
The general approach to generating random variates through transformations which are not one-to-one is discussed. A known transformation of a random variable with an inverse Gaussian distribution to one which follows the chi-square distribution with one degree-of-freedom provides an example of the ambiguity involved when a transformation does not have a single-valued inverse. The general approach and special properties of the inverse Gaussian yield a new and useful technique for the generation of variates from the inverse Gaussian distribution. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 26, 1974
- Accession Number
- AD0782676
Entities
People
- John R. Michael
- Roy W. Haas
- William R. Schucany
Organizations
- Southern Methodist University