## ＜プレプリント＞Notes on Estimating Inverse-Gaussian and Gamma Subordinators under High-frequency Sampling

作成者 著者識別子 作成者名 所属機関 所属機関名 Graduate School of Mathematics, Kyushu University 九州大学大学院数理学研究院 英語 Faculty of Mathematics, Kyushu University 九州大学大学院数理学研究院 2006-09-14 2006-28 Author's Original open access We study joint efficient estimation of two parameters dominating gamma and inverse-Gaussian subordinators, based on discrete observations sampled at $(t^n_ i)^n_{i=1}$ satisfying $h_n := max_{i leq... n}(t^n_i - t^n_{i-1}) \rigtarrow 0$ as $n \rightarrow infty$. Under the condition that $T_n := t^n_n \rigtarrow infty$ as $n \rightarrow infty$ we have two kinds of optimal rates, $sqrt{n}$ and $sqrt{T_n}$, and especially. Moreover, as in estimation of diffusion coefficient of a Wiener process the $sqrt{n}$-consistent component of the estimator is effectively workable even when $T_n$ does not tend to infinity. Simulation experiments are given under several $h_n$’s behaviors続きを見る

### 本文情報を非表示

2006-28 pdf 263 KB 90

### 詳細

レコードID 3395 査読無 MHF Preprint Series || 2006-28 || p1-9 http://www.math.kyushu-u.ac.jp/gakufu/ Kyushu University 21st Century COE Program Development of Dynamic Mathematics with High Functionality 九州大学21世紀COEプログラム「機能数理学の構築と展開」 プレプリント 2009.04.22 2018.02.26