Abstract:

This paper presents the representation of an asset pricing model rule assuming the absence of arbitrage, the existence of market frictions, and the Put-Call Parity. This model constitutes a special case of the Choquet Pricing Rule, where the non-additive probability measure (or capacity) is decomposed into an additive probability and an increasing weighting function. The necessary conditions for a Choquet Pricing Rule to be a Rank-Dependent Pricing Rule are given in the finite and the infinite cases. We test the empirical validity of the Put-Call and Call-Put Parities assumptions on Bid and Ask call option prices from the S&P500. The Rank-Dependent Pricing Rule is calibrated on the same data, utilizing two new families of distortion functions tailored for flexibility and two other functions referred to as (Generalized) Neo-Additive Capacity. We investigate the impact of time to expiration (time value) and moneyness (intrinsic value) on the shape of the distortion function. The resulting models always exhibit a greater accuracy than the benchmark (linear model). Furthermore, the calibrated distortion functions display a remarkably similar shape. The results from the calibration procedure allow us, through the inverted S-shape distortion function, to conclude the risk-seeking behaviour of the market in evaluating call options prices. Finally, we verify the robustness of the calibration on another dataset.

Location:

Room 2E30
ENS Paris-Saclay, 4 avenue des Sciences, 91190, Gif-sur-Yvette