Investigating the Impact of Technical Indicators on Option Price Prediction through Deep Learning Models

Time series analysis and financial market forecasting are two areas that lend themselves well to technical analysis, which uses historical values as indicators to reveal possible market patterns. The development of a prediction model that faithfully represents the trend of monetary activity is very desirable since it allows for more educated choices. This study examines the effect of technical indicators on option price prediction using three deep learning models: Long Short Term Memory, Convolutional Neural Network, and Gated Recurrent Unit. As input characteristics for the models, a carefully selected set of eleven technical indicators is used with S&P500 option fundamental data. The evaluation of implemented models is conducted by comparing their performances utilizing established evaluation metrics, namely Mean Absolute Percentage Error (MAPE), Determination Coefficient (R2), Mean Absolute Error(MAE), and Root Mean Square Error (RMSE). Evidence suggests that a model’s predicting performance can be improved even more by feeding it a combination of indicators.