文摘
An asymptotic theory was given by Phillips and Magdalinos (J Econom 136(1):115鈥?30, 2007) for autoregressive time series Y t 鈥?鈥?em class="a-plus-plus">蟻Y t鈭?鈥?鈥?em class="a-plus-plus">u t , t鈥?鈥?,...,n, with 蟻鈥?鈥?em class="a-plus-plus">蟻 n 鈥?鈥?鈥?鈥?em class="a-plus-plus">c/k n , under (2鈥?鈥?em class="a-plus-plus">未)-order moment condition for the innovations u t , where 未鈥?gt;鈥? when c鈥?lt;鈥? and 未鈥?鈥? when c鈥?gt;鈥?, {u t } is a sequence of independent and identically distributed random variables, and (k n ) n鈥夆垐鈥夆剷 is a deterministic sequence increasing to infinity at a rate slower than n. In the present paper, we established similar results when the truncated second moment of the innovations $l(x)=\textsf{E} [u_1^2I\{|u_1|\le x\}]$ is a slowly varying function at 鈭? which may tend to infinity as x鈥夆啋鈥夆垶. More interestingly, we proposed a new pivotal for the coefficient 蟻 in case c鈥?lt;鈥?, and formally proved that it has an asymptotically standard normal distribution and is nuisance parameter free. Our numerical simulation results show that the distribution of this pivotal approximates the standard normal distribution well under normal innovations.