Gaussian Estimation of Parametric Spectral Density with Unknown Pole

Gaussian Estimation of Parametric Spectral Density with Unknown Pole
复制标题

DOI:
10.1214/aos/1013699989
复制
发表时间:
2001-08
期刊:
Econometrics: Econometric & Statistical Methods - General eJournal
影响因子:
--
通讯作者:
L. Giraitis;J. Hidalgo
L. Giraitis;J. Hidalgo
中科院分区:
其他
文献类型:
--
作者:
L. Giraitis;J. Hidalgo

文献摘要

被引文献

相似文献

我们考虑一个参数谱密度在未知频率w上的分数极具有幂律行为。未知w的情况,特别是w = 0,在长记忆文献中是标准的。当w未知时,参数估计的渐近分布理论,包括(长)记忆参数,是非常困难的。我们研究了高斯估计的一种形式。我们建立了w的估计的n相合性,并讨论了它的(非标准)极限分布性质。对于剩余的参数估计,我们建立了vn一致性和渐近正态性。
We consider a parametric spectral density with power-law behaviour about a fractional pole at the unknown frequency w. The case of unknown w, especially w = 0, is standard in the long memory literature. When w is unknown, asymptotic distribution theory for estimates of parameters, including the (long) memory parameter, is significantly harder. We study a form of Gaussian estimate. We establsih n-consistency of the estimate of w, and discuss its (non-standard) limiting distributional behaviour. For the remaining parameter estimates, we establish Vn-consistency and asymptotic normality.