Extreme Value Theory in Applied Probability.
Abstract
Several problems of applied probability lead to the investigation of extremes of a sequence of random variables (floods, strength of materials, failure models, maximal queue length, air pollution data, etc.). It is unrealistic in most cases to assume that the random variables in question are independent and identically distributed. However, there are several dependent models when the distribution of the k-th largest is well approximated by the distribution of the k-th largest of independent, and some times independent and identically distributed random variables (possibly changing the fixed size of the sequence in question to a random one).
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1978
- Accession Number
- ADA059021
Entities
People
- Janos Galambos
Organizations
- Temple University