riskcore-cpp

High-performance C++ equity risk analytics engine with real-time WebSocket streaming and live web dashboard. Compute 95% Value-at-Risk, Black-Scholes Greeks, and portfolio metrics with sub-millisecond latency.

Overview

riskcore-cpp delivers institutional-grade portfolio risk analytics:

• 95% Historical VaR (1-day, per position and portfolio)
• Black-Scholes Greeks (delta, gamma, vega, theta for ATM options)
• Sharpe Ratio (annualized with 4.5% risk-free rate)
• Variance-Covariance Portfolio VaR with Pearson correlation matrix
• Real-time Dashboard with 4 interactive chart panels and live metrics

System Architecture

graph LR
    A["Market Data<br/>yfinance"] --> B["Data Loader<br/>CSV/JSON"]
    B --> C["Risk Engine<br/>C++20"]
    C --> D["WebSocket Server<br/>BSD Sockets"]
    D --> E["Web Dashboard<br/>Chart.js"]
    C --> F["JSON Output<br/>--run mode"]

    style C fill:#0099cc
    style D fill:#00cc88
    style E fill:#ff5555

Loading

Computation Pipeline

sequenceDiagram
    participant Market as Market Data
    participant Engine as Risk Engine
    participant Calc as Calculations
    participant WS as WebSocket
    participant UI as Dashboard

    Market->>Engine: Load positions + returns
    Engine->>Calc: VaR, Greeks, Sharpe, Correlation
    Calc->>Engine: Results (2ms typical)
    Engine->>WS: Broadcast JSON
    WS->>UI: WebSocket frame
    UI->>UI: Update 4 charts + metrics

Loading

Features

Feature	Details
Risk Metrics	VaR 95%, Greeks, Sharpe, correlation, P&L tracking
Performance	Sub-millisecond computation, 60-tick rolling history
Real-time Updates	2-second refresh cycle via WebSocket
Interactive UI	Dark theme, responsive layout, 4 live charts
Data Pipeline	yfinance → CSV/JSON → C++ engine
Portability	C++20, zero external math libraries, STL only

Quick Start

Prerequisites

brew install cmake libwebsockets pkg-config openssl
curl -LsSf https://astral.sh/uv/install.sh | sh

Build & Run

git clone https://github.com/nim444/riskcore-cpp.git && cd riskcore-cpp

# Fetch market data
uv run scripts/fetch_data.py

# Build
cmake -B build -DCMAKE_BUILD_TYPE=Release
cmake --build build --parallel

# Single computation (JSON output)
./build/riskcore --run

# Live dashboard (2 terminals):
# Terminal 1:
./build/riskcore --serve

# Terminal 2:
python3 -m http.server 8000 --directory web
# Open http://localhost:8000

Dashboard Panels

1. Portfolio VaR Trend — 60-point rolling window with dynamic Y-axis zoom
2. Sharpe Rolling — Real-time ratio with 0.5 target reference line
3. VaR Heatmap — Per-ticker risk sorted by impact (4-tier colour contrast)
4. P&L Analysis — Dollar or percentage mode toggle (all 8 positions readable)

Technology Stack

Layer	Technology
Core	C++20, STL algorithms, BSD sockets
Build	CMake 3.20+, clang++ (Apple Silicon)
Backend	libwebsockets, OpenSSL, nlohmann/json
Frontend	Vanilla JS, Chart.js, CSS3
Data	Python, yfinance, uv

Configuration

Custom Tickers

Edit scripts/fetch_data.py:

tickers = ["IBM", "GOOG", "NVDA", "MSFT", "AAPL", "TSLA", "AMZN", "META", "YOUR_TICKER"]
position_config = {"YOUR_TICKER": {"side": "LONG", "quantity": 1000}}

Then: uv run scripts/fetch_data.py && cmake --build build --parallel && ./build/riskcore --serve

CLI Reference

./build/riskcore --run       # Compute once, output JSON
./build/riskcore --serve     # Start WebSocket server (port 8080)
./build/riskcore --version   # Show version

Documentation

• Distributed Architecture Extension — Design for multi-node risk computation and federated analytics

Troubleshooting

Issue	Solution
WebSocket won't connect	Ensure ./build/riskcore --serve is running on port 8080
Dashboard shows "undefined"	Verify web server runs on port 8000
Port conflict	lsof -i :8080 to find process
$0 prices	Run uv run scripts/fetch_data.py to fetch data

License

MIT License – See LICENSE file