Please wait a minute...
Applied Mathematics A Journal of Chinese Universities  2017, Vol. 32 Issue (1): 41-48    DOI:
    
Composite quantile estimation for moderate deviations from a unit root model with possibly infinite variance errors
NI Jia-lin, FU Ke-ang
School of Stat. and Math., Zhejiang Gongshang Univ., Hangzhou 310018, China
Download:
Export: BibTeX | EndNote (RIS)      

Abstract  Under the mildly integrated and the mildly explosive cases, the asymptotic distributions of composite quantile estimation for moderate deviations from a unit root model with possibly infinite variance errors are obtained, respectively. Some simulation studies are also given to show that the composite quantile estimation has a good performance.

Key wordsautoregression      unit root      domain of attraction of the normal law      heavy tail      composite quantile estimation     
Received: 02 August 2016      Published: 18 March 2018
CLC:  O212.1  
Cite this article:

NI Jia-lin, FU Ke-ang. Composite quantile estimation for moderate deviations from a unit root model with possibly infinite variance errors. Applied Mathematics A Journal of Chinese Universities, 2017, 32(1): 41-48.

URL:

http://www.zjujournals.com/amjcua/     OR     http://www.zjujournals.com/amjcua/Y2017/V32/I1/41


方差可能无穷的“中度偏离”单位根过程的复合分位数估计

考虑一类“中度偏离”单位根过程,$y_t = q_ny_{t-1}+u_t$,其中$q_n=1+\frac{c}{k_n}$,$k_n=o(n)$,$c$为一非零常数,$\{u_t\}$为随机扰动项序列. 在允许扰动项方差无穷的条件下,构造$q_n$的复合分位数估计, 并得到了该估计的渐近分布. 最后通过数值模拟, 在扰动项服从$t(2)$分布下, 说明了该估计的稳健和有效性.

关键词: 自回归,  单位根,  正态吸引场,  重尾,  复合分位数 
[1] YUAN Qiao-li, WU Liu-cang, DAI Lin. Parameters estimation for mixture of double generalized linear models [J]. Applied Mathematics A Journal of Chinese Universities, 2017, 32(3): 267-276.