Asymptotic Distribution Expansions in Option Pricing

This article extends an option pricing model and studies the properties of implied probability distribution functions (PDFs) recovered from American interest rate futures options. The model relaxes the restricting assumption of lognormally distributed returns accommodating a wide variety of implied PDFs shapes. Non-normal skewness and kurtosis are found to contribute significantly to estimating more precise implied densities. The model achieves good in-sample accuracy, similar to that achieved by alternative approaches. The recovered implied PDFs are, finally, found to be closer to a median PDF estimated using a number of alternative techniques