Computation of a joint distribution for scan statistics with temporal-spatial structures

2016 Fiscal Year Final Research Report

• Project/Area Number: 24500356
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: The Institute of Statistical Mathematics
• Principal Investigator: Kuriki Satoshi 統計数理研究所, 数理・推論研究系, 教授 (90195545)
• Co-Investigator(Renkei-kenkyūsha): TAKAHASHI Kunihiko 名古屋大学, 大学院・医学系研究科, 准教授 (50323259) ; HARA Hisayuki 同志社大学, 文化情報学部, 准教授 (40312988)
• Project Period (FY): 2012-04-01 – 2017-03-31
• Keywords: 空間疫学 / 多重性調整 / マルコフ性 / コーダルグラフ

Outline of Final Research Achievements

By extracting the Markov structure from a set of random variables with temporal and spatial structures, we have established a fast computation method to evaluate the multiplicity-adjusted p-value of temporal-spatial scan statistics. In addition, we applied the proposed method to spatial epidemiology. We demonstrated the usefulness of the proposed method by analyzing several data. The basic idea is (i) to describe the geographic information (adjacent relationship) by a non-directed graph, and (ii) to extract the Markov structure via the chordal extension of the graph, and (iii) to evaluate the multiplicity-adjusted p-value by successive numerical integrations based on the Markov property. On Saturday 29th October 2016, at the Institute of Statistical Mathematics (Tachikawa, Tokyo), we organized the workshop "New Horizons of Applied Statistics".

Free Research Field

統計科学