A Nonparametric Method for Estimating Interaction Effect of Age and Period on Mortality

• Published in: Environmental health perspectives Vol. 87; pp. 115 - 121
• Main Authors: Ohtaki, Megu, Kim, Dong-Kyu, Munaka, Masaki
• Format: Journal Article
• Language: English
• Published: United States National Institute of Environmental Health Sciences. National Institutes of Health. Department of Health, Education and Welfare 01.07.1990
• Subjects: Age Factors; Age groups; Analysis of Variance; Anus; Anus Neoplasms - mortality; Application of Nonparametric Smoothing; Cancer; Cohort Studies; Data smoothing; Estimation methods; Female; Humans; Japan - epidemiology; Legends; Male; Modeling; Mortality; Parametric models; Rectal Neoplasms - mortality; Rectum; Sigmoid Neoplasms - mortality; Time Factors; Japan
• ISSN: 0091-6765; 1552-9924
• DOI: 10.1289/ehp.9087115

In this paper we introduce a new model and develop an estimation strategy to analyze mortality data. The model we dealt with has the specific structure E[ log qij]=μ +αi+βj+ρijsubject to the linear restrictions ∑tαi/σi 2=∑jβj=∑ipij/σi 2=∑jρij=0 for any i and j; here, qij denotes the mortality for the jth period category and ith age category, μ denotes the overall mean, αidenotes the ith age effect in antichronological order, βjthe jth period effect, ρijthe general interaction effect; and σi 2is the common error variance of log qijfor the ith age group. We propose a combined technique of ANOVA and nonparametric smoothing for estimating these parameters. The methods described are illustrated by mortality data on rectum cancer in Japanese males and females between 1950-1986.