Point-Figure Stock Back testing

I want to quantify how my point-and-figure rules perform on a set of individual equities. I will supply the daily price history in a simple CSV format (or you can fetch it yourself through yfinance, Quandl, etc.). Your task is to build a clean, reproducible back-test that lets me tweak box-size and reversal parameters and immediately see updated results.

At the end of each run I need the script to return:
• Win/Loss ratio
• Return on Investment (ROI)
• Max Drawdown

Python is preferred because I already work with pandas, NumPy and Jupyter, but I am open to R or a platform such as AmiBroker if full source code and step-by-step instructions come with the delivery. A modular design that separates data import, point-and-figure plotting, signal generation and performance reporting will make future tweaks easier.

The back-test should:
– Process roughly 100 tickers over 10-20 years of daily data without choking on memory or speed.
– Produce a concise trade log and a summary CSV with the three metrics above.
– Match trades to conventional point-and-figure charts on random spot checks; accuracy in the plotting routine is essential.

Deliverables
• Well-commented source code or notebook
• Brief README explaining setup and how to change parameters
• Sample output files (trade log and metrics summary)

Once this baseline is working we can explore position sizing, portfolio-level analytics and additional entry filters, but the first milestone is a solid, repeatable back-test returning those three core metrics.