Docs / nablatensor-quant / com.nablatensor.quant
final class
SabrHagan
Hagan et al. (2002) lognormal-implied-volatility approximation for SABR.
Two forms: a plain double double, double, double, double, double, double) for generating reference quotes, and a tape-level ADouble, ADouble, ADouble, ADouble, double, double, double) whose alpha, beta, rho and nu are differentiable — the model function inside Calibrator.
Methods
static double blackVol(double alpha, double beta, double rho, double nu, double forward, double strike, double maturity)
Reference (host) Black vol.
static ADouble blackVol(AadRecorder rec, ADouble alpha, ADouble beta, ADouble rho, ADouble nu, double forward, double strike, double maturity)
Tape-level Black vol. forward, strike and maturity are constants; the four SABR parameters are differentiable. Requires strike != forward (use a small offset for the ATM point).