A1317
Title: Confidence interval construction: A new self-normalization approach based on adjusted range
Authors: Yongmiao Hong - Department of Economics at Cornell University (United States)
Oliver Linton - University of Cambridge (United Kingdom)
Jiajing Sun - University of Chinese Academy of Sciences (China) [presenting]
Shouyang Wang - Academy of Mathematics and Systems Science, Chinese Academy of Sciences (China)
Abstract: A new self-normalization method is proposed to construct confidence intervals for quantities of stationary time series. Unlike the self-normalization approach, which utilizes the variance of a partial as the self-normalizer, we propose the use of its adjusted range instead. Given ranges capacity to deal long-range dependence in highly non-Gaussian time series with large skewness and/or kurtosis, we introduce two range-based autocorrelation tests and study a confidence interval construction for censored dependent data extending previous work. The simulations confirm the validity of our approach.