Risk Measurement and Margin Systems of Equity Derivatives’ Portfolio: Theory and Evidence
博士 === 國立中山大學 === 財務管理學系研究所 === 102 === We modified the popular SPAN margining model with a diagonal model to construct a margining system called Beta-Simulation. We use Beta-Simulation to calculate margin requirements for portfolios that include stock index futures contract, stocks, and stock optio...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2014
|
Online Access: | http://ndltd.ncl.edu.tw/handle/24396568437289379180 |
Summary: | 博士 === 國立中山大學 === 財務管理學系研究所 === 102 === We modified the popular SPAN margining model with a diagonal model to construct a margining system called Beta-Simulation. We use Beta-Simulation to calculate margin requirements for portfolios that include stock index futures contract, stocks, and stock options. The Beta-Simulation system uses historical stock beta to simplify setting the appropriate requirements for collateral offset estimates for inter-commodity spreads. Our model performs computational procedure more easily than SPAN and offers a sounder theoretical basis than TIMS for credit offset estimates among individual stock options.
The Beta-Simulation model back tests competing systems using data from U.S. markets. The test results show that the new margining system provides the same market risk protection as the SPAN system but with collateral levels that are substantially less than those required by SPAN. Other popular margining systems for stock options TIMS cannot provide the same coverage for U.S. market data. Therefore, the Beta-Simulation system is shown to be a better model of calculating margins requirements and for measuring risk with respect to portfolios containing stock derivatives.
We also test the Beta-Simulation model empirically using all actual open positions by the Taiwan Futures Exchange’s clearing members. The back tests show that the new model requires only 74% of the TAIFEX SPAN margin requirements to offer the same protection at the same confidence interval. Our model also passes the simulated stress test by being assessed against the period when the financial tsunami swept the world.
|
---|