A simple two-step method for spatio-temporal design-based balanced sampling

作     者：Khavarzadeh, Ramin Mohammadzadeh, Mohsen Mateu, Jorge 

作者机构：Tarbiat Modares Univ Dept Stat Tehran Iran Univ Jaume 1 Dept Math Castellon de La Plana Spain 

出 版 物：《STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT》 (随机环境研究与风险评估)

年 卷 期：2018年第32卷第2期

页      面：457-468页

核心收录：

学科分类：0830[工学-环境科学与工程（可授工学、理学、农学学位）] 08[工学] 081501[工学-水文学及水资源] 0815[工学-水利工程] 0714[理学-统计学（可授理学、经济学学位）] 0814[工学-土木工程] 

基　　金：Ordered and Spatial Data Center of Excellence of the Ferdowsi University of Mashhad 

主　　题：Balanced sampling Design-based sampling Spatio-temporal sampling 

摘      要：We introduce a two-step method to perform spatio-temporal balanced sampling in a design-based approach. For populations with spatio-temporal trends and with anisotropic effects in the variable of interest, the prediction can be further improved by selecting samples that are well spread over the entire population in space and time. We control the spread of the sample over the population by using the volume of the corresponding three-dimensional Voronoi tessellation. Indeed, spatio-temporal design-based balanced sampling is even more efficient under the presence of a trend and anisotropic effects. We present an intensive simulation study comparing our method to other available methods for spatio-temporal sampling. Finally, we analyze real data by sampling from a population of temperature stations over six European countries.