Option Pricing with Time-Varying Volatility Risk Aversion

TL;DR

This paper develops a new option pricing model incorporating time-varying volatility risk aversion, linking the variance risk ratio to economic fundamentals and sentiment, and demonstrates improved pricing accuracy empirically.

Contribution

It introduces a tractable model combining a time-varying risk aversion kernel with the Heston-Nandi GARCH framework, providing analytical formulas and empirical validation.

Findings

Significant reduction in pricing errors for S&P 500 options.

Variance risk ratio closely linked to economic and sentiment measures.

Analytical approximation method enhances computational efficiency.

Abstract

We introduce a pricing kernel with time-varying volatility risk aversion to explain observed time variations in the shape of the pricing kernel. When combined with the Heston-Nandi GARCH model, this framework yields a tractable option pricing model in which the variance risk ratio (VRR) emerges as a key variable. We show that the VRR is closely linked to economic fundamentals, as well as sentiment and uncertainty measures. A novel approximation method provides analytical option pricing formulas, and we demonstrate substantial reductions in pricing errors through an empirical application to the S&P 500 index, the CBOE VIX, and option prices.