On Multiple-hypotheses-testing and Model Selection

• Activity Number: 57
• Type: Contributed
• Date/Time: Sunday, August 8, 2004 : 4:00 PM to 5:50 PM
• Sponsor: Business and Economics Statistics Section
• Author(s): Kasing Man*+ and Chung Chen
• Companies: Syracuse University and Syracuse University
• Keywords: time series ; model selection ; vector ARMA model

Abstract

It is of interest to study the presence or the absence of specific dynamic relations between time series. Chen and Lee (1990) proposed a multiple-hypotheses-testing procedure in a vector ARMA framework to identify the possible dynamic relation (which include independent, contemporaneous relation, unidirectional relation, and feedback relation) between time series. This paper extends this procedure to account for strong form relationship. It is a sequential inference procedure based on the likelihood ratio tests on models with various parametric constraints. On the other hand, viewing models with different parametric constraints as different models, they can be compared using model selection criteria such as AIC and BIC. It will be interesting to see how the two approaches compare and relate to each other. Simulation studies indicate that the performance of AIC is similar to the procedure conducting at a higher significance level; while BIC at a lower significance level. For empirical analysis, we study some economic time series to illustrate the procedures. Robustness of the results over possible alternative model specification will be addressed.