Fast and accurate computation of the exact confidence limits for the common odds ratio in several 2 x 2 tables

Jiangang (Jason) Liao, Charles B. Hall

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Mehta, Patel, and Gray proposed a network algorithm for the exact confidence limits of the common odds ratio in several 2x2 tables. Their algorithm was implemented in the StatXact and Egret statistical packages and further discussed by Vollset, Hirji, and Elashoff. The need to evaluate polynomials of potentially very high degrees, however, poses some numerical difficulties. This article presents a method that cuts the degree of polynomials by at least one half. Two other modifications to further stabilize and speed the computation are also proposed.

Original languageEnglish (US)
Pages (from-to)173-179
Number of pages7
JournalJournal of Computational and Graphical Statistics
Volume4
Issue number3
DOIs
StatePublished - Jan 1 1995

Fingerprint

Common ratio
Confidence Limits
Odds Ratio
Tables
Statistical package
Polynomial
Network Algorithms
Evaluate
Odds ratio
Polynomials
Confidence

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Discrete Mathematics and Combinatorics
  • Statistics, Probability and Uncertainty

Cite this

@article{b30fe7823d4e4784961ecd2c2213ffc4,
title = "Fast and accurate computation of the exact confidence limits for the common odds ratio in several 2 x 2 tables",
abstract = "Mehta, Patel, and Gray proposed a network algorithm for the exact confidence limits of the common odds ratio in several 2x2 tables. Their algorithm was implemented in the StatXact and Egret statistical packages and further discussed by Vollset, Hirji, and Elashoff. The need to evaluate polynomials of potentially very high degrees, however, poses some numerical difficulties. This article presents a method that cuts the degree of polynomials by at least one half. Two other modifications to further stabilize and speed the computation are also proposed.",
author = "Liao, {Jiangang (Jason)} and Hall, {Charles B.}",
year = "1995",
month = "1",
day = "1",
doi = "10.1080/10618600.1995.10474675",
language = "English (US)",
volume = "4",
pages = "173--179",
journal = "Journal of Computational and Graphical Statistics",
issn = "1061-8600",
publisher = "American Statistical Association",
number = "3",

}

TY - JOUR

T1 - Fast and accurate computation of the exact confidence limits for the common odds ratio in several 2 x 2 tables

AU - Liao, Jiangang (Jason)

AU - Hall, Charles B.

PY - 1995/1/1

Y1 - 1995/1/1

N2 - Mehta, Patel, and Gray proposed a network algorithm for the exact confidence limits of the common odds ratio in several 2x2 tables. Their algorithm was implemented in the StatXact and Egret statistical packages and further discussed by Vollset, Hirji, and Elashoff. The need to evaluate polynomials of potentially very high degrees, however, poses some numerical difficulties. This article presents a method that cuts the degree of polynomials by at least one half. Two other modifications to further stabilize and speed the computation are also proposed.

AB - Mehta, Patel, and Gray proposed a network algorithm for the exact confidence limits of the common odds ratio in several 2x2 tables. Their algorithm was implemented in the StatXact and Egret statistical packages and further discussed by Vollset, Hirji, and Elashoff. The need to evaluate polynomials of potentially very high degrees, however, poses some numerical difficulties. This article presents a method that cuts the degree of polynomials by at least one half. Two other modifications to further stabilize and speed the computation are also proposed.

UR - http://www.scopus.com/inward/record.url?scp=0344159729&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0344159729&partnerID=8YFLogxK

U2 - 10.1080/10618600.1995.10474675

DO - 10.1080/10618600.1995.10474675

M3 - Article

AN - SCOPUS:0344159729

VL - 4

SP - 173

EP - 179

JO - Journal of Computational and Graphical Statistics

JF - Journal of Computational and Graphical Statistics

SN - 1061-8600

IS - 3

ER -