Stock Trend Prediction Using Deep Learning Approach on Technical Indicator and Industrial Specific Information

A stock trend prediction has been in the spotlight from past to present. Fortunately, there is an enormous amount of information available nowadays. There were prior attempts that have tried forecast using textual information; however, it can be further improved since they relied on fixed word embedding, and depends sentiment whole market. In this paper, we propose a deep learning model predict Thailand Futures Exchange (TFEX) with ability analyze both numerical information. We used Thai economic news headlines various online sources. To obtain better sentiment, divided into industry-specific indexes (also called “sectors”) reflect movement securities same fundamental. The proposed method consists Long Short-Term Memory Network (LSTM) Bidirectional Encoder Representations Transformers (BERT) architectures daily market activity. evaluated performance by considering predictive accuracy returns obtained simulation buying selling. experimental results demonstrate enhancing each sector improve outperform all baselines.