Technical analysis-based unsupervised intraday trading djia index stocks: is it profitable in long term?

Abstract

The paradigm shift from conventional stock market trading rings to computer-driven algorithmic trading has given rise to a new era characterized by specialized trading systems and indicators meticulously engineered to decode price charts and enhance the prospects of profitable trading. Nevertheless, despite these notable advancements, the majority of traders continue to grapple with losses rather than realizing gains, echoing the historical pursuit of the elusive philosopher’s stone by alchemists of yore. In response to this challenge, our research delves into the realm of artificial neural networks (ANNs) to cultivate more sophisticated trading methodologies. Our empirical investigations suggest that trading strategies relying on price chart analysis generally achieve a moderate level of accuracy. However, it is imperative to acknowledge that the intricate patterns that materialize over time, coupled with return metrics, persistently elude precise prediction within the framework of unsupervised automated trading. These findings underscore the critical importance of embracing a progressive approach to trading that synergizes human expertise with cutting-edge technological capabilities.