The three-Group Causal Path applied to VECM
Granger-Causality in VAR model is discussed generally by partition all variables into two groups. However, in some cases, a causal path with more than two groups is just practical enough. The goal of this paper is considering a three group system with the following three groups: 1) independent group, 2)intermediate group, and 3)dependent group and applies such a causality in the VECM model.A multiple hypothesis testing is proposed for testing the existence of three-group causality when the parameters are unknown in the VECM representation. Finally, we apply such a multiple hypothesis testing to a case study about the causality between property insurance industry and macroeconomic variables in Taiwan, and it can be shown that property insurance industry will influence (lead) macroeconomic variables, but not vice versa.
Three-Group Causal Path Vector Autoregressive Process (VAR) Vector Error Correction Model(VECM) ultiple Hypothesis Testing Property Insurance Industry
Neng-Fang, Tseng
Department of Statistics and Actuarial Science Aletheia University Taiwan
国际会议
The 10th International Conference on Intelligent Technologies(第十届智慧科技国际会议 InTech09)
桂林
英文
252-256
2009-12-12（万方平台首次上网日期，不代表论文的发表时间）