Cohort effect specification for evaluating cancer risk

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K08740
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Epidemiology and preventive medicine
• Research Institution: Sapporo Medical University
• Principal Investigator: Kamo Kenichi 札幌医科大学, 医療人育成センター, 准教授 (10404740)
• Co-Investigator(Kenkyū-buntansha): 佐藤 健一 広島大学, 原爆放射線医科学研究所, 准教授 (30284219) ; 冨田 哲治 県立広島大学, 経営情報学部, 准教授 (60346533)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: がん / 生まれ年効果 / APCモデル / 識別問題 / ポアソン回帰

Outline of Final Research Achievements

For the longitudinal cancer risk, three main elements are "age", "period" and "birth cohort." The APC (age-period-cohort) model handle these at the same time. However, there are linear dependent relationship among these elements. Then we can not estimate each element without some constraint. In our reserch project, we note how to specify birth cohort effect automatically by using statistical model.

Free Research Field

数学