Implications of Parameter Uncertainty on Option Prices

摘要

Financial markets are complex processes where investors interactto set prices. We present a framework for option valuation under imperfectinformation, taking risk neutral parameter uncertainty into account. Theframework is a direct generalization of the existing valuation methodology. Many investors base their decisions on mathematical models that have beencalibrated to market prices. We argue that the calibration process introducesa source of uncertainty that needs to be taken into account. The modelsand parameters used may differ to such extent that one investor may find anoption underpriced; whereas another investor may find the very same optionoverpriced. This problem is not taken into account by any of the standardmodels. The paper is concluded by presenting simulations and an empirical studyon FX options, where we demonstrate improved predictive performance (in sample and out of sample) using this framework.