BondYield-Py

Educational bond price and yield calculator with a C++ core, pybind11 bindings, and PySide6 GUI.

BondYield-Py is a desktop application for calculating fixed-rate bond prices and yields. The project combines a C++ bond math engine with a Python/PySide6 graphical user interface using pybind11 bindings.

The application demonstrates two different yield calculation techniques and highlights the trade-offs between simplicity, convergence speed, and supported input ranges.

Features

• Bond price calculation
• Yield calculation using a stepwise search method
• Yield calculation using Deeley's method
• Calculation statistics including iteration count and execution time
• Input validation and error handling
• Automated pytest test suite
• C++ calculation engine exposed to Python through pybind11
• Cross-platform architecture using CMake and PySide6

Architecture

PySide6 GUI
     ↓
Python Package
     ↓
pybind11 Bindings
     ↓
C++ Bond Math Engine

The bond pricing and yield calculation algorithms are implemented in C++ while the user interface is implemented in Python using PySide6.

Requirements

• Python 3.10 or later
• CMake 3.20 or later
• C++20 compatible compiler

Currently tested on:

• macOS (Apple Silicon)
• Windows 11
• Linux

Installation

Installation builds the C++ extension module and installs all required Python dependencies automatically.

Clone the repository:

git clone git@github.com:gconde/BondYield-Py.git
cd BondYield-Py

macOS / Linux

Create and activate a virtual environment:

python -m venv .venv
source .venv/bin/activate

Install the package:

pip install -e .

Windows

Open:

x64 Native Tools Command Prompt for VS 2022

The x64 Native Tools prompt is important because the C++ extension must be built with a 64-bit compiler to match 64-bit Python.

Then run:

python -m venv .venv
.venv\Scripts\activate
pip install -e .

Running Tests

Install the development dependencies:

pip install -e ".[dev]"

Run the test suite:

pytest

Running the Application

bond-yield

Bond Model Assumptions

This demonstration application uses the following assumptions:

• Fixed-rate coupon bonds
• One coupon payment per period
• Integer periods to maturity
• Coupon rates expressed as decimal values (0.05 = 5%)
• Yield rates expressed as decimal values (0.05 = 5%)

The application is intended as an educational demonstration rather than a production fixed-income analytics library.

Yield Calculation Methods

Stepwise Search

The stepwise search method begins with an approximate yield estimate and iteratively adjusts the rate until the calculated bond price matches the target price within a specified tolerance.

Advantages:

• Simple implementation
• Supports a wider range of valid inputs

Disadvantages:

• May require many iterations to converge
• Slower than Deeley's method

If convergence is not achieved within the iteration limit, the method returns a yield value of -1.

Deeley's Method

Deeley's method uses a more sophisticated iterative approach that generally converges much faster than the stepwise search method.

Advantages:

• Significantly fewer iterations
• Faster convergence for supported inputs

Disadvantages:

• Supports a narrower range of inputs in this demonstration

For unsupported inputs, the GUI reports that Deeley's method is unavailable while still allowing the stepwise search method to be used.

Supported Deeley Input Range

The Deeley implementation has been tested for the following input ranges:

• Coupon rate ≥ 1%
• Periods to maturity ≥ 2
• Purchase price between 50% and 100% of face value

Inputs outside these ranges may still be processed successfully by the stepwise search method.

Testing

The test suite includes:

• Bond pricing validation
• Yield repricing consistency tests
• Yield method comparison tests
• Invalid input validation tests
• Non-finite value (NaN and infinity) validation tests
• Edge case regression tests

References

Chris Deeley, Superseding Newton with a Superior Bond Yield Algorithm

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1253166