作者: Yixiao Sun , Jingjing Yang
DOI: 10.1016/J.JECONOM.2020.06.007
关键词: Heteroscedasticity 、 Variance (accounting) 、 Quadratic equation 、 Autocorrelation 、 Applied mathematics 、 Point estimation 、 Kernel (statistics) 、 Mathematics 、 Type I and type II errors 、 Inference
摘要: Abstract The paper investigates the optimal kernel choice in heteroskedasticity and autocorrelation robust tests based on fixed-b asymptotics. In parallel with optimality of quadratic spectral under asymptotic mean squared error criterion point estimator long run variance as considered Andrews (1991) , we show that continues to hold testing-oriented Sun, Philips Jin (2008) which takes a weighted average probabilities type I II errors test.