This paper reviews recent advancements in machine learning (ML) driven automated trading systems (ATS). ATS has progressed from simple rule-based systems to sophisticated ML models like deep reinforcement learning, deep learning, and Q-learning that can adapt to evolving markets. These techniques have been successfully applied across various financial instruments to optimize trading strategies, forecast prices, and enhance profits. The literature indicates that ML improves ATS performance over conventional methods by identifying intricate patterns and relationships in data. However, risks like overfitting, instability, and low interpretability exist. Techniques to mitigate these limitations include cross-validation, careful model management, and utilizing more transparent algorithms. Although challenges remain, ML creates valuable opportunities for ATS via alternative data sources, advanced feature engineering, optimized adaptive strategies, and holistic market modelling. While research shows ML improves market quality through increased liquidity and efficiency, heightened volatility needs further analysis. Promising future research directions include leveraging innovations in deep learning, reinforcement learning, sentiment analysis, and hybrid systems. More work is also needed on evaluating different techniques systematically. Overall, the progress in ML-driven ATS contributes significantly to the field, but judicious application and balanced regulations are required to address risks. Further advancements in ML will enable more capable, nuanced, and profitable algorithmic trading.

This paper analyses the performance of machine learning models in forecasting the Tehran Stock Exchange's automobile index. Historical daily data from 2018-2022 was pre-processed and used to train Linear Regression (LR), Support Vector Regression (SVR), and Random Forest (RF) models. The models were evaluated on mean absolute error, mean squared error, root mean squared error and R2 score metrics. The results indicate that LR and SVR outperformed RF in predicting automobile stock prices, with LR achieving the lowest error scores. This demonstrates the capability of machine learning techniques to model complex, nonlinear relationships in financial time series data. This pioneering study on a previously unexplored dataset provides empirical evidence that LR and SVR can reliably forecast automobile stock market prices, holding promise for investing applications.