This repository contains a Python script that fetches historical Bitcoin (BTC) price data, calculates technical indicators, and applies a trading strategy based on machine learning. The purpose of the script is to assess the performance of the strategy using the XGBoost algorithm, which predicts long, short, or neutral positions.

Implemented a trading strategy using XgBoost to create a predictive model on the direction and variations of the BTC/USDT pair.

Autonomous trading bot for Polymarket prediction markets using dual strategies: XGBoost ML models (60% WR) + technical indicator confluence. Trades BTC/ETH 5min/15min Up/Down markets with intelligent bet sizing and risk management.

BTC trading strategy using XGBoost with buy/sell signals

Engineered an end-to-end backtesting framework to validate a machine learning trading strategy for BTC/USDT. Developed a data pipeline from the Binance API, trained an XGBoost model, and simulated trades with transaction costs. The system quantified the strategy's performance, revealing a Sharpe Ratio of -12.9 and preventing potential capital loss.