Deep ReLU network expression rates for option prices in high-dimensional, exponential Lévy models

We study the expression rates of deep neural networks (DNNs for short) for option prices written on baskets of $d$ d risky assets whose log-returns are modelled by a multivariate Lévy process with general correlation structure of jumps. We establish sufficient conditions on the characteristic triplet of the Lévy process $X$ X that ensure $\varepsilon $ ε error of DNN expressed option prices with DNNs of size that grows polynomially with respect to ${\mathcal{O}}(\varepsilon ^{-1})$ O ( ε − 1 ) , and with constants implied in ${\mathcal{O}}(\, \cdot \, )$ O ( ⋅ ) which grow polynomially in $d$ d , thereby overcoming the curse of dimensionality (CoD) and justifying the use of DNNs in financial modelling of large baskets in markets with jumps.In addition, we exploit parabolic smoothing of Kolmogorov partial integro-differential equations for certain multivariate Lévy processes to present alternative architectures of ReLU (“rectified linear unit”) DNNs that provide $\varepsilon $ ε expression error in DNN size ${\mathcal{O}}(|\log (\varepsilon )|^{a})$ O ( | log ( ε ) | a ) with exponent $a$ a proportional to $d$ d , but with constants implied in ${\mathcal{O}}(\, \cdot \, )$ O ( ⋅ ) growing exponentially with respect to $d$ d . Under stronger, dimension-uniform non-degeneracy conditions on the Lévy symbol, we obtain algebraic expression rates of option prices in exponential Lévy models which are free from the curse of dimensionality. In this case, the ReLU DNN expression rates of prices depend on certain sparsity conditions on the characteristic Lévy triplet. We indicate several consequences and possible extensions of the presented results.

Regime Classification and Stock Loan Valuation

For traditional perpetual American put options under regime-switching models with positive risk-free interest rates, optimal stopping usually can occur in any regime. Nonetheless, if the risk-free interest rates are allowed to equal zero (the interest rate may drop to zero sometimes in reality), there may exist “continuation regimes” within which optimal stopping can never occur, that is, within which stopping is never optimal. A natural problem is “regime classification,” that is, determination of all continuation regimes. In “Regime Classification and Stock Loan Valuation,” Ning Cai and Wei Zhang develop a unified, fixed point approach to solving this regime classification problem under general regime-switching exponential Levy models with any finite numbers of regimes and general Levy types. Applying this result, they also provide a unified framework for the valuation of infinite maturity stock loans under general regime-switching exponential Levy models.

SWING OPTION PRICING BY DYNAMIC PROGRAMMING WITH B-SPLINE DENSITY PROJECTION

Swing options are a type of exotic financial derivative which generalize American options to allow for multiple early-exercise actions during the contract period. These contracts are widely traded in commodity and energy markets, but are often difficult to value using standard techniques due to their complexity and strong path-dependency. There are numerous interesting varieties of swing options, which differ in terms of their intermediate cash flows, and the constraints (both local and global) which they impose on early-exercise (swing) decisions. We introduce an efficient and general purpose transform-based method for pricing discrete and continuously monitored swing options under exponential Lévy models, which applies to contracts with fixed rights clauses, as well as recovery time delays between exercise. The approach combines dynamic programming with an efficient method for calculating the continuation value between monitoring dates, and applies generally to multiple early-exercise contracts, providing a unified framework for pricing a large class of exotic derivatives. Efficiency and accuracy of the method are supported by a series of numerical experiments which further provide benchmark prices for future research.