在Black-Scholes公式中,波动率σ是一个非常重要的参数.并且在诸如股票、利率、股指期货等标的资产的交易市场中,人们往往希望知道标的资产未来价格的波动率,从而知道该资产的未来风险结构.但一般来说,由于事件还没有发生,人们对σ的未来走向很难预测.但可以运用Black-Scholes的理论框架,从期权市场获取的信息去重构标的资产价格的波动率.论文使用的是基于Tikhonov正则化的数值微分方法,利用Dupire公式去重构标的资产的未来预期波动率.相对于其他方法,该算法更加快速有效,并且能识别标的资产的预期风险突变。
In Black-Scholes model, a is an important parameter, which is also not freely observable in the market. The volatility of the future prices of the underlying assets which can be stocks, interest, futures and so on, is a wonder. When one knows the volatility, risk structure of those assets is in hand. It turns out that in practice, it is difficult to forecast the future trends of underlying assets directly. Using Black-Scholes equation, this paper recovers the local volatility of underlying assets from information in the option market. In this paper, based on Tikhonov regulation method, numerical differentiation is used to recover the volatility from Dupire formula. Compared with other methods ,this method is faster and more stable. The outstanding advantage of this method is that it can recognize the risk saltation of underlying assets.