ABSTRACT

This chapter studies inference of implied parameters of different financial models from observed prices. In particular, it considers inference of a pair (σimp(t), ρimp(t)) of two unconditionally implied parameters, where σimp(t) is the unconditionally implied volatility, and ρimp(t) is the unconditionally implied value of ρ(t). The basic pricing rule for models with random volatility is risk-neutral valuation, when the option price is given as the expected value of its future payoff with respect to a risk-neutral measure discounted back to the present time t. The chapter presents an example of a Markovian setting, when maximization over a class of volatilities can be reduced to the solution of some nonlinear parabolic equations. It examines estimation of the future cumulative short-term interest rate and the related problem of bond pricing.