From Data to Decisions: A Trading Bot for Ethereum

Abstract

This research focuses on the development and evaluation of an automated trading strategy for cryptocurrency markets, specifically Ethereum, using Pine Script version 5 on the Trading View platform. The strategy incorporates a variety of technical indicators, including the Relative Strength Index (RSI), Commodity Channel Index (CCI), Average True Range (ATR), Directional Movement Index (DMI), Aroon Indicator, and Exponential Moving Average (EMA), to capture diverse market conditions such as trend direction, volatility, and overbought/oversold levels. To enhance decision-making, a multiple logistic regression model was applied, allowing the trading bot to predict market movements by assigning probabilities to different outcomes based on the interplay of these indicators. The strategy was tested on a 3-minute timeframe for both long and short positions between January 6 and January 22, 2025, with an initial capital of 271, with 319 closed trades. The win rate of 22.88% was complemented by a risk-reward ratio of 1.3, emphasizing the strategy’s ability to maximize returns from successful trades. Key performance metrics, including a Sharpe ratio of 0.4 and a Sortino ratio of 1.067, highlighted the strategy’s relative stability and ability to handle downside risk in volatile cryptocurrency markets. The research demonstrates the effectiveness of integrating multiple technical indicators with logistic regression in automated trading systems, offering valuable insights into the potential of algorithmic trading in highly volatile environments like the cryptocurrency market.