Release History
Initial release (2026-09-28)
This first documented release of quantplus introduces the package as a small,
performance-oriented option-pricing library built around the Cox-Ross-Rubinstein
(CRR) binomial model.
Included features
CRR option pricing for European and American options
Black-Scholes closed-form pricing for benchmarking
tree-output utilities for inspecting terminal stock and option states
optional stock-price threshold pruning at expiry or every tree step
an interactive tree plot with draggable thresholds and live repricing
Cython-based Python bindings around a native C++ implementation
regression tests covering correctness and convergence behavior
Project goals
The package is intentionally focused on a narrow but important problem: pricing vanilla options with transparent numerical methods and a simple API. The design is meant to be easy for users to understand while still leveraging native code for performance-critical computations.
Future direction
Possible future additions include:
more option types and payoff structures
more advanced Greeks calculations
additional numerical benchmarks and validation examples
additional documentation and usage examples for quantitative finance workflows
The current release is deliberately small and readable, making it a solid base for future extension without increasing conceptual complexity.