1. Consistent Testing for Pairwise Dependence in Time Series
- Author
-
Fokianos, Konstantinos, Pitsillou, M., and Fokianos, Konstantinos [0000-0002-0051-711X]
- Subjects
Statistics and Probability ,Time series ,Covariance function ,Generalized spectral density ,01 natural sciences ,Order of integration ,Normal distribution ,010104 statistics & probability ,Spectral density ,Statistical tests ,V-statistics ,Joint probability distribution ,V-statistic ,0502 economics and business ,Statistics ,Test statistic ,Applied mathematics ,Random variables ,0101 mathematics ,Statistic ,050205 econometrics ,Mathematics ,Distance covariance ,Applied Mathematics ,05 social sciences ,U-statistic ,Distance correlation ,Empirical characteristic function ,Kernel ,Modeling and Simulation ,U-statistics ,Random variable - Abstract
We consider the problem of testing pairwise dependence for stationary time series. For this, we suggest the use of a Box–Ljung-type test statistic that is formed after calculating the distance covariance function among pairs of observations. The distance covariance function is a suitable measure for detecting dependencies between observations as it is based on the distance between the characteristic function of the joint distribution of the random variables and the product of the marginals. We show that, under the null hypothesis of independence and under mild regularity conditions, the test statistic converges to a normal random variable. The results are complemented by several examples. This article has supplementary material online. © 2017 American Statistical Association and the American Society for Quality. 59 2 262 270
- Published
- 2017
- Full Text
- View/download PDF