2014 Fiscal Year Final Research Report
Research on Statistical Modeling and Methods for Quantitative Risk Management
| Project/Area Number |
24500347
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | Seijo University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2012-04-01 – 2015-03-31
|
| Keywords | リスク管理 / 計量ファイナンス / 統計的手法 |
|---|
| Outline of Final Research Achievements |
The class of distortion risk measures is a wide class with many desirable properties, and includes the renowned expected shortfall. We constructed a reasonable and natural estimator of distortion risk measure based on general weakly dependent times series data, and proved that it is strongly consistent and asymptotically normal. Furthermore we gave a consistent estimator for its asymptotic variance, and discussed bias correction methods using bootstrap methods. We also suggested a simple backtesting procedure for the distortion risk measuress, and computed the Euler capital allocation based on them with some numerical implementation.
We proposed a new version of smoothed empirical copula, called the empirical Beta copula. It has the advantage of not requiring any smoothing parameter, and it is extremely simple to simulate a sample from it. We showed asymptotic results on the empirical Beta copula and studied its finite-sample properties with Monte Carlo simulation.
|
| Free Research Field |
統計科学
|
