Publication result detail

Statistical Correlation in Stratified Sampling

VOŘECHOVSKÝ, M., NOVÁK, D.

Original Title

Statistical Correlation in Stratified Sampling

English Title

Statistical Correlation in Stratified Sampling

Type

Paper in proceedings (conference paper)

Original Abstract

Clanek predstavuje novou metodu pro zavedeni statisticke zavislosti mezi vzorky samplovane metodami tyou Monte Carlo

English abstract

A new efficient technique to impose the statistical correlation when using Monte Carlo type method for statistical analysis of computational problems is proposed. The technique is based on stochastic optimization method called Simulated Annealing. The comparison with other techniques presently used and intensive numerical testing showed the superiority and robustness of the method. No significant obstacles have been found working also with large problems (large number of random variables). Advantages and limitations of the approach will be discussed. Remarks on positive definiteness of target correlation matrix are made. Numerical examples show the efficiency of the method.

Key words in English

correlation, Monte Carlo simulation, Latin hypercube sampling, statistical analysis, sensitivity, reliability, stochastic optimization, Simulated Annealing

Authors

VOŘECHOVSKÝ, M., NOVÁK, D.

Released

06.07.2003

Publisher

Cerra

Location

San Francisco

ISBN

90-5966-004-9

Book

ICASP 9, The Ninth International Conference on Applications of Statistics and Probability in Civil Engineering

Pages from

91

Pages count

7

BibTex

@inproceedings{BUT13920,
  author="Miroslav {Vořechovský} and Drahomír {Novák}",
  title="Statistical Correlation in Stratified Sampling",
  booktitle="ICASP 9, The Ninth International Conference on Applications of Statistics and Probability in Civil Engineering",
  year="2003",
  pages="7",
  publisher="Cerra",
  address="San Francisco",
  isbn="90-5966-004-9"
}