Quantum

A lightweight IDE for quants. Write a small script in Quantum's own language (.qtm files), hit Run, and get live charts, comparisons, and Monte Carlo risk simulations rendered straight into the editor - no Jupyter notebook required.

Why

Most quant workflows live in notebooks: re-running cells, re-importing yfinance, copy-pasting boilerplate just to look at a price chart. Quantum strips that down to one line per action:

load(AMZN)
display()
compare(AAPL)

Press Run. That's it.

Features

• Custom scripting language (.qtm) - one command per line, plain English-ish syntax, no Python boilerplate required to use it
• Dracula-themed code editor with syntax highlighting and line numbers
• Live charts rendered inline via matplotlib, driven by yfinance
• Side-by-side comparison of two tickers
• Live-updating watch mode that polls a ticker on a timer without freezing the UI
• Monte Carlo portfolio simulation with VaR/CVaR across four models (historical, normal, t-distribution, and simulated)
• Console log with its own syntax highlighting for quick scanning

Commands

See docs/functions.md for full syntax and options. Quick reference:

Command	What it does
load(TICKER, RANGE)	Fetch price history, set the active ticker
display()	Render a chart for the active ticker
compare(TICKER, RANGE)	Render a second chart alongside the first
range(RANGE)	Change the time range for everything after it
watch(TICKER, SECONDS, MINUTES)	Live-refreshing chart on a timer
unwatch()	Stop the live chart
monte(TICKER+TICKER, options...)	Monte Carlo VaR/CVaR simulation

Getting started

Requirements

• Qt 6 with the MinGW toolchain
• Python 3 with yfinance, matplotlib, pandas, numpy, and scipy installed (pip install yfinance matplotlib pandas numpy scipy)

Build

build.ps1 expects Qt installed at C:\Qt\6.11.1\mingw_64 and the MinGW compiler at C:\Qt\Tools\mingw1310_64 - adjust the paths at the top of the script if your install differs.

.\build.ps1

This regenerates the Qt moc files, compiles everything, and launches the app against test_script.qtm.

Project layout

src/            C++ source (Qt widgets, syntax highlighter, command parser)
scripts/        Python scripts the app shells out to (yfinance, matplotlib)
resources/      Logo, button icons, and generated chart images (gitignored)
docs/           Language reference and notes
test_script.qtm Example script used by build.ps1 on launch

Architecture

The C++ side is a thin shell: a QPlainTextEdit-based code editor with a custom gutter (CodeEditor), a syntax highlighter for both the editor and the console (SyntaxHighlighter), and a CommandHandler that owns all script logic. Adding a new command means adding one case in CommandHandler::execute() and a handleX() method - MainWindow.cpp never needs to change.

Each command shells out to a dedicated Python script (scripts/fetchPrice.py, scripts/watchPrice.py, scripts/montecarlo.py), which does the actual data fetching and plotting, then writes a PNG that the C++ side loads into a QLabel. watch() runs its script asynchronously via a persistent QProcess so the UI never blocks waiting on a live refresh.