Option pricing and portfolio choice

Aramonte, Sirio (2009) Option pricing and portfolio choice. Doctoral thesis, University of London: London Business School.

Abstract

The focus of this dissertation is on option pricing, in particular on the economic determinants of option risk premia, and on the interaction between investor sentiment and innovation in the context of technology diffusion. The rest of the three chapters empirically investigates whether macroeconomic uncertainty is a priced risk factor in the cross-section of equity option returns. The analysis employs a factor model, estimated with the Fama-MacBeth methodology, and the macroeconomic uncertainty factor is based on options' "excess" pricing errors on days immediately before scheduled macroeconomic announcements. I find that macroeconomic uncertainty is priced in the cross-section of equity option returns, even after controlling for a large set of relevant factors. In addition, introducing the macroeconomic uncertainty factor affects the estimated risk premia on the market, the volatility of volatility and the volatility of jumps. The results are robust to measurement error in stock and option prices, to possible biases generated by the non-linearity of option returns and by non-randomly missing returns, and to several methods of measuring macroeconomic uncertainty and expected volatility. The second chapter studies how the interaction between technological innovation and investor sentiment affects firm-level investment and aggregate productivity growth. When firms face the decision to adopt a new technology with uncertain productivity, small scale experimentation is a direct way of obtaining information useful to evaluate full-scale adoption. If such information is not appropriable, a free-rider problem arises and the aggregate level of investment is sub-optimal. I hypothesize that investor sentiment mitigates the effect of informational externalities, which would make investor sentiment socially valuable in the context of technology diffusion. I find that investor sentiment increases the effect of technological innovation on investment for firms more susceptible to informational externalities, and that investor sentiment also raises the impact of technological innovation on productivity growth. The third chapter focuses on portfolio choice, investigating whether distributions implied in option prices can help reducing estimation error for expected returns, given that the cross-section of option prices contains information about the moments of expected returns that is not available in the underlying's prices. I study the performance of portfolio strategies based on option-implied distributions relative to that of the robust 1/N allocation rule, and I find that option-based models outperform the 1/N strategy, especially with (limited) short-selling. Transaction costs are relatively high for option-based strategies, and they reduce, but do not eliminate, the economic significance of the results.