WEKO3
インデックスリンク
アイテム
Detecting Structural Changes in Stochastic Differential Equation System Based Upon a Bayesian Approach
http://hdl.handle.net/11316/00000763
http://hdl.handle.net/11316/000007634027b5d4-4a21-42b6-9800-270e972b67fc
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
本文（Article） (1.7 MB)
|
|
| Item type | 紀要論文 / Departmental Bulletin Paper(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 2018-07-26 | |||||
| タイトル | ||||||
| タイトル | Detecting Structural Changes in Stochastic Differential Equation System Based Upon a Bayesian Approach | |||||
| 言語 | ||||||
| 言語 | eng | |||||
| 資源タイプ | ||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||
| 資源タイプ | departmental bulletin paper | |||||
| その他（別言語等）のタイトル | ||||||
| その他のタイトル | Detecting Structural Changes in Stochastic Differential Equation System Based Upon a Bayesian Approach | |||||
| 著者 |
譚, 康融
× 譚, 康融 |
|||||
| 抄録 | ||||||
| 内容記述タイプ | Abstract | |||||
| 内容記述 | In recent years, researches have been carried out to investigate the evolving of the Stochastic Differential Equations（SDEs）, which are utilized for describing many scientific phenomena, such as, the growth of population, the consumption of natural gas, the behavior of stock prices and returns. One of the most famous models in finance is to use a SDE, the Geometric Brownian Motion （GBM） to depict the ups and downs of the prices（returns）. Furthermore, some studies suggested jump factor should be plugged into the above GBM model, Merton’s jump model, for example. This paper deals with the detecting of structural changes in such SDEs. From the previous studies, we know that there are several parameters in the above mentioned SDEs which determine the evolving values, for instance, drift, diffusion, and jump term etc. The structural changes will occur while the parameters in the SDE change. Equivalently, the evolution of SDE will differ as soon as change points of parameters present. So far, many studies on detecting structural changes have been carried out, though, the most of them need the assumption of normality. However, for change point or structural change problem, normality seldom holds in many cases. In this paper, we propose a Bayesian approach to detect where a structural change exactly occurs. Our proposed approach is to firstly consider the change sizes of the evolving values（such as prices, returns）, secondly to deal with these counting numbers of the change sizes during the fixed intervals as a poisson process, and then by using a Gibbs sampler to detect where the change point lies at. Through our numerical analysis, we find out that our proposed approach works well. | |||||
| 書誌情報 |
経済社会研究 en : he journal of the Society for Studies on Economies and Societies 巻 58, 号 1-2, p. 51-67, 発行日 2018-06-25 |
|||||
| 出版者 | ||||||
| 出版者 | 久留米大学経済社会研究会 | |||||
| ISSN | ||||||
| 収録物識別子タイプ | ISSN | |||||
| 収録物識別子 | 2433-2682 | |||||
| 書誌レコードID（NCID） | ||||||
| 収録物識別子タイプ | NCID | |||||
| 収録物識別子 | AA12584414 | |||||
