2017 Fiscal Year Final Research Report
A study of statistical methods in quantitative risk management, focusing on copulas and risk measures
| Project/Area Number |
15H03337
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Economic statistics
|
|---|
| Research Institution | Seijo University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
川崎 能典 統計数理研究所, モデリング研究系, 教授 (70249910)
|
|---|
| Research Collaborator |
SEGERS Johan
MCNEIL Alex J.
EMBRECHTYS Paul
MITTNIK Stefan
|
|---|
| Project Period (FY) |
2015-04-01 – 2018-03-31
|
| Keywords | 定量的リスク管理 / リスク計測手法 / リスク尺度 / 接合関数 / コピュラ / ボラティリティ / リスク因子の探索 / 非対称分布のモデリング |
|---|
| Outline of Final Research Achievements |
Financial institutions are faced with various kinds of financial risk by holding their positions. Quantitative risk management deals with how to identify, evaluate, measure and manage those risks. We discussed the statistical properties that desirable risk measures should possess, and studied a broad class of risk measures called distortion risk measures focusing on their statistical estimation and backtesting. Copulas are an important tool for modeling dependence between several financial risks, and we introduced and analyzed new smoothed estimates of copulas. Resampling methods based on them are also investigated.
|
| Free Research Field |
統計学・計量ファイナンス
|
