简体繁体 English

QuantLib 中的隐含波动率是否独立于定价引擎？

[英]Is the implied volatility in QuantLib independent of the pricing engine?

原文 2020-06-26 15:28:20 1 1 quantitative-finance/ quantlib/ volatility

This might be a stupid question as I just started looking into QuantLib.当我刚开始研究 QuantLib 时，这可能是一个愚蠢的问题。 (I'm using the python API.) It seems that the implied volatility calculation does not require a pricing engine. （我使用的是 python API。）似乎隐含波动率计算不需要定价引擎。 If so, which pricing engine is used?如果是这样，使用哪个定价引擎？ I am interested in the implied volatility of American options.我对美式期权的隐含波动率感兴趣。

1 个解决方案

The problem here is that the option cannot reuse the same engine set to the instrument.这里的问题是该选项无法将相同的引擎集重新用于仪器。 This is because it needs to create a new engine, in which the volatility is flat and under the control of the method (since it needs to be changed by the solver).这是因为它需要创建一个新引擎，其中波动率是平坦的并且在方法的控制下（因为它需要由求解器更改）。 This can't be done in a generic way, especially considering that users might implement entirely new engines.这不能以通用方式完成，尤其是考虑到用户可能会实现全新的引擎。

All the method can do is return an estimate based on one chosen engine.该方法所能做的就是根据一个选择的引擎返回一个估计值。 The default chosen is FdBlackScholesVanillaEngine with default parameters.默认选择的是FdBlackScholesVanillaEngine和默认参数。 If you want to use a different engine, or the same engine with different parameters, you'll have to duplicate the code inside the method.如果您想使用不同的引擎，或具有不同参数的相同引擎，则必须在方法内复制代码。