题　目：A Permutation Test for Cluster Detection 主讲人：余清祥教授（政治大学统计学系） 时　间：95年11月3日(星期五)上午10:40-11:30 (上午10:20-10:40茶会於交大统计所428室举行) 地　点：交大综合一馆427室 Abstract In recent years, many statistical methods have been proposed for detecting excesses of rare diseases, i.e., clusters, in space or in space-time. Most of these methods deal with case-event or individual-level data and are designed to detect clusters with shape close to circles. In this study, adapting the idea of Choynowski (1957), a two-stage simulation-based approach is proposed to detect both circular and non-circular clusters with aggregate or group-level data. The proposed cluster-detection method will be used to compare with Spatial scan statistic (Kulldorff, 1995) and Flexible scan statistic (Tango, 2005), two frequently used methods. Computer simulation is used to illustrate the validity of the proposed approach. In addition, the cancer mortality and fertility data of Taiwan area are also used to demonstrate the proposed test. Key words. Cluster detection, Rare disease, Aggregate data, Monte Carlo Simulation, Permutation Test. --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 140.113.114.186 ※ catherinejoy:转录至看板 NCTU-STAT94G 10/19 00:57