This bibliometric analysis offers a comprehensive overview of the evolving landscape of Operations Research (OR) applications in financial and stock market forecasting. This review systematically extracted and analysed 722 articles published between 2015 and 2026, following a PRISMA-guided two-stage screening of 1,495 records retrieved from Scopus. The findings reveal significant trends and insights in the OR-forecasting literature. The study finds that the field is strongly interdisciplinary, with research activity concentrated around LSTM, neural network, SVM, ARIMA, and CNN methods. The keyword and thematic analysis uncovered clusters spanning sentiment analysis, natural language processing, and technical analysis, alongside the growing use of machine learning and deep learning in forecasting applications. The findings suggest emerging gaps around the systematic comparison of forecasting techniques and the integration of structured and unstructured data. These insights offer valuable guidance for researchers and practitioners seeking to develop hybrid, standardised, and reproducible forecasting frameworks, and highlight opportunities for greater methodological integration between OR and AI-driven finance.