Title: Random weighting the Portmanteau tests for multivariate white noise with unknown dependent structure
Authors: Yanfen Zhang - Xiamen University (China) [presenting]
Abstract: The random weighting method is extended from univariate to multivariate to bootstrap the critical value of Ljung-Box portmanteau test in multivariate weakly white noise. A set of Monte Carlo experiments that check multivariate white noise illustrate the practical relevance of this method. A real example on the USD-MYR and USD-SGD five day rate of return on foreign exchange illustrates the merits of our testing procedure.