PIRanNN: Random-Feature Pricing and Calibration under Jump-Diffusion and Stochastic-Volatility Models
Organizations: Department of Mathematical Sciences, Rajiv Gandhi Institute of Petroleum Technology, Jais, Amethi 229304, Uttar Pradesh, India · Department of Management Studies, Indian Institute of Science, Bengaluru 560012, Karnataka, India · Department of Mathematics, Indian Institute of Science, Bengaluru 560012, Karnataka, India
Abstract
We study a random-feature method for pricing and calibrating European options under jump-diffusion (Merton, Kou) and stochastic-volatility (Heston, SABR) models. The network has one hidden layer with a probit activation and fixed random inner weights, so fitting it to the pricing equation, the payoff and the boundary conditions is a single regularised least-squares problem, with no iterative training. For Merton and Kou the jump integral of each feature is explicit, so no quadrature is needed. One factorisation of the least-squares matrix gives the price at every state and time, its Greeks, and the exact derivatives of the prices with respect to the model parameters for gradient-based calibration. For the jump models, subtracting the Black-Scholes price removes the payoff kink and reduces the pricing error by about two orders of magnitude. The method is then about as accurate as a standard Crank-Nicolson finite-difference scheme at the same run time, although a finite-difference scheme that also subtracts the Black-Scholes price is about two orders of magnitude more accurate still. On the test problems of a recent physics-informed neural network study, the method matches that network's accuracy in seconds instead of about half an hour. The calibration experiments show that a solver can fit the quotes and still return wrong parameters when its pricing error is not small compared with the effect of the parameters on the prices: without the subtraction the solver misses the jump parameters by up to 50%, and with it they are recovered to within 3%. For Heston and SABR, used as an applicability study with boundary conditions that need no reference prices, pricing errors are about one percent and the calibrated parameters are within about 15% of the true values.
Figures & tables
| Inverse PINN | Parametric PINN / operator network | PIRanNN (this paper) | |
|---|---|---|---|
| What is fitted | network weights and model parameters together | network weights, offline, over a parameter box | output weights, at each trial parameter vector |
| Offline cost | none | large | none |
| Price at a trial parameter vector | not computed separately | evaluation of the trained surrogate | one regularised least-squares solve |
| Derivative in the parameters | implicit in the joint training | automatic differentiation of the surrogate | exact; reuses the factorisation of the price |
| Parameter range | unrestricted | limited to the training box | unrestricted |
| Fitting the network | nonconvex | nonconvex (offline) | convex (least squares) |
| Symbol | Meaning | Symbol | Meaning |
|---|---|---|---|
| log-moneyness | standard normal distribution function and density | ||
| time to maturity | number of random features | ||
| price in units of the strike | inner weight vector and centre of feature | ||
| normalised call payoff | pre-activation | ||
| pricing operator | output weights | ||
| expected value after a jump | numbers of interior, initial, boundary points |
| Experiment | Model parameters | Domain | ||
|---|---|---|---|---|
| Merton pricing | , , , , , | , ; finite differences on | ||
| Kou pricing | , , , , , , | , ; finite differences on | ||
| Jump calibration | as for pricing | , | ( in the repeat study) | |
| PINN comparison | as in Bansal et al. (2026) | , | , , | (median) |
| Heston | Cases 1–4 of In ’t Hout and Foulon (2010) (Table 10 ) | , , (pricing), (calibration) | (pricing), (calibration) | |
| SABR | , , , , | , , (pricing), (calibration), (Monte Carlo comparison) | (pricing), (otherwise) |
| Maturity | overall | ||||
|---|---|---|---|---|---|
| Merton (exact reference) | |||||
| PIRanNN , plain | |||||
| PIRanNN , with subtraction | |||||
| FDM, explicit jump term | |||||
| FDM, implicit jump term | |||||
| FDM, implicit, with subtraction | |||||
| Configuration | Merton | Kou |
|---|---|---|
| neither ingredient | ||
| extra features near the strike | ||
| Black–Scholes subtraction | ||
| both |
| Merton | Kou | ||||
|---|---|---|---|---|---|
| Jump term | rel. RMSE | time (s) | Jump term | rel. RMSE | time (s) |
| exact | exact | ||||
| Merton | Kou | |||||||
| Model | Parameter | True | Start | Plain solve | With subtraction | ||
| Merton | |||||||
| pricing error at the true parameters | |||||||
| price RMSE, start | |||||||
| At the money | Strikes to | |||||
|---|---|---|---|---|---|---|
| Case | Plain | With subtraction | Plain | With subtraction | ||
| 1 | ( ) | ( ) | ||||
| 2 | ( ) | ( ) | ||||
| 3 | ( ) | ( ) | ||||
| 4 | ( ) | ( ) | ||||
| Method | Single draw | Price averaged over draws |
|---|---|---|
| Hagan’s formula | ||
| PIRanNN , plain solve | ( to ) | |
| PIRanNN , with subtraction | ( to ) | |
| Model | Parameter | True | Recovered | Rel. error |
|---|---|---|---|---|
| Heston | ||||
| SABR |
Appendix figures & tables1 asset
Supplementary material from the paper’s appendix.
Appendix
| Parameter | Value | Plain solve | With subtraction |
|---|---|---|---|
| Intensity | |||
| Jump volatility | |||
Quantitative Finance
Quantitative finance uses mathematical models, statistics, and machine learning to value assets, measure risk, and guide trading and portfolio decisions.
See all 208 papers