Trading Auto-bot for Enhanced Financial Decision Making

This paper develops an automated trading system based on the backtesting of strategies on the Trader’s View platform, which simulates multiple market scenarios. Machine learning models KNN and KNN with PSO is used for backtesting to measure the performance based on risk-adjusted returns, drawdown, and profit consistency. These selections were also incorporated into the Zerodha API using the KNN-PSO model, which can connect with real markets for the automated execution of trades and showed a considerable outperformance in return on investment and risk management. The model in real time was greatly comparable with what had been in the back-tested result, thus evolving accordingly with changing market conditions. This work is promising specially in terms of short term trading for the improvement of profitability and decision-making as an insight for individual traders and institutions and proved by comparing with LSTM and KNN-PSO ensemble model.