DerivaSharp.Gpu.Linux 1.0.0

DerivaSharp

DerivaSharp is a .NET 10 library for pricing single-asset derivatives under the Black-Scholes-Merton model. It provides a consistent, strongly typed API across closed-form, numerical integration, finite-difference, binomial-tree, and Monte Carlo engines.

The library is intended for quantitative research, model validation, prototyping, and integration into .NET pricing services.

Capabilities

Additional functionality includes:

• Price and risk measures through a common engine API: delta, gamma, speed, theta, charm, color, vega, vanna, zomma, and rho.
• Implied-volatility and implied-coupon solvers.
• Vectorized price, delta, and gamma calculations over spot grids.
• CPU execution by default, with optional CUDA acceleration for Monte Carlo engines.
• Actual/365 Fixed day counts, a null calendar, and an SSE trading calendar.
• Numerical utilities for quadrature, root finding, interpolation, distributions, and tridiagonal systems.

Requirements

• .NET 10 SDK
• Windows x64 or Linux x64 when using an optional GPU runtime package
• For GPU execution, a CUDA-capable NVIDIA GPU and compatible driver

Installation

Release packages use the following NuGet IDs:

dotnet add package DerivaSharp

For CUDA 12.8 support, add the platform-specific companion package:

dotnet add package DerivaSharp.Gpu.Windows
# or
dotnet add package DerivaSharp.Gpu.Linux

Quick Start

The following example prices a one-year European call and calculates its analytic Greeks:

using DerivaSharp.Instruments;
using DerivaSharp.Models;
using DerivaSharp.PricingEngines;
using DerivaSharp.Time;

DateOnly valuationDate = new(2025, 1, 6);
DateOnly expirationDate = valuationDate.AddYears(1);

EuropeanOption option = new(
    OptionType.Call,
    strikePrice: 100.0,
    effectiveDate: valuationDate,
    expirationDate);

BsmModelParameters model = new(
    volatility: 0.30,
    riskFreeRate: 0.04,
    dividendYield: 0.01);

PricingContext<BsmModelParameters> context = new(
    model,
    AssetPrice: 100.0,
    ValuationDate: valuationDate,
    Calendar: NullCalendar.Shared);

AnalyticEuropeanEngine engine = new();
PricingResult result = engine.ValueAndGreeks(option, context);

Console.WriteLine($"PV:    {result.Value:F6}"); // 13.151137
Console.WriteLine($"Delta: {result.Delta:F6}"); // 0.592749
Console.WriteLine($"Gamma: {result.Gamma:F6}"); // 0.012761

Engines are interchangeable where the instrument type permits it:

FdEuropeanEngine engine = new(
    FiniteDifferenceScheme.CrankNicolson,
    priceStepCount: 1_000,
    timeStepCount: 1_000);

double presentValue = engine.Value(option, context);
double impliedVolatility = engine.ImpliedVolatility(option, context, optionPrice: 13.151137);

Monte Carlo engines accept a deterministic seed and can opt into CUDA:

McEuropeanEngine engine = new(
    pathCount: 500_000,
    stepCount: 2,
    useCuda: true,
    seed: 42);

Conventions

Build and Validate

dotnet restore DerivaSharp.slnx
dotnet build DerivaSharp.slnx --no-restore
dotnet test DerivaSharp.slnx --no-build --verbosity normal

Run the BenchmarkDotNet suite in Release mode:

dotnet run --project benchmarks/DerivaSharp.Benchmarks.csproj --configuration Release

Interactive examples are available in notebooks/, including European-option, accumulator, and snowball workflows.