A CONSERVATIVE USE OF HISTORICAL DATA FOR A TREND TEST IN PROPORTIONS

Takashi Yanagawa; David G. Hoel; Gary T. Brooks

doi:10.11329/jjss1970.19.83

Takashi Yanagawa, David G. Hoel, Gary T. Brooks

著者情報

キーワード: Beta-binomial distribution, Carcinogenicity bioassay, Cochran-Armitage test, Historical control, Hoel-Yanagawa test

ジャーナルフリー

1989 年 19 巻 1 号 p. 83-94

DOI https://doi.org/10.11329/jjss1970.19.83

詳細

抄録

The exact conditional tests for incorporating historical controls in testing for a trend in proportions, which were developed recently by Hoel and Yanagawa (1986), depend on Beta-binomial parameters which are estimated from historical data. This paper considers the conditional test for logistic response by Hoel and Yanagawa, shows the test is sensitive to this source of variability, and discusses a conservative use of the test.

責任著者(Corresponding author)

J-STAGEへの登録はこちら（無料）