A Portfolio Optimization Model For Banks Based On Monte Carlo Simulation And The Constraint Of VaR Technology

The level of asset-liability portfolio management has sizable influence on the market competitiveness and profit ability of bank. With the increasing by a wide margin of company's bankrupt phenomenon in the global range, the credit risks have already become one of the main risks that banks facing, especially like this to Chinese commercial bank. And then, carrying on effective measurement and control to the credit risks, and the closely related optimization question of the bank's assets portfolio become the focus of both academia and banking at present. It has important realistic meanings to study the optimization model of asset-liability portfolio and rational distribution of assets structure which suitable to Chinese commercial banks.Based on Monte Carlo simulation and the CreditMetrics method, considering the constrain on VaR, laws, regulations, and operation, using portfolio profits maximum of bank's assets as objective function, the optimal model of asset-liability management is set up in this paper, in order to provides decision-making method for bank's risk management.The characteristics of this thesis lies on four aspects: Firstly, It makes the data needed more reasonable by using Monte Carlo simulation to forecast the earning rates year-by-year and then figure up the average earning rates and mean square deviations. Secondly, the risk of loan distribution is limited within given ranges of bank's risk tolerance ability and reserve funds, because default risk of loan portfolio is controlled by the arrangement on using VaR constrain and maximum limitation of loss under certain confidence. Thirdly, liquidity risk is controlled by using constrains on laws, regulation and operation, so the loan' allocation can meet the requirement of supervision and operation. Finally, the loan's yields of historical data on individual enterprise are used to get the correlation coefficient between different loans, and get portfolio deviation, thus, the yield's correlation among different loans is reflected directly.