Price action analysis in gold trading using deep learning techniques

We introduce a new structural deep learning model with the purpose of learning price action trading features that subjective traders use to make trading decisions based on recent and actual price movements, rather than relying solely on technical indicators. The model combines Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to improve trend forecasting for gold prices and generate better trading signals compared to traditional strategies. Since gold prices are represented as time series data, it is appropriate to apply CNN and LSTM for forecasting. Our model concept revolves around CNN’s ability to detect price action features or patterns at different locations within the time series data, while LSTM maintains both short-term and long-term memory as a sequence over time. By leveraging their collaborative capabilities, the neural network model comprehends complex relationships between recent and actual price movements and trends in gold prices. Our study demonstrates that the combination of CNN and LSTM with price action trading features significantly enhances long-term trading performance.