Financial Temporal Patterns and Deep FinanicalNet For Crude-Oil Prices Forecasting

Time series forecasting has very important role in time variant systems such as financial markets, investors and prediction. Predicting volatile behavior of crude oil time series data is a challenging task because it depends upon not only the existing nature of a system but also external factors such as environmental conditions and economic status of a system. This paper presents insights into financial variable West Texas Intermediate (WTI) crude oil price investigations. We introduced temporal financial pattern to encode the volatile behavior of crude oil time series data. Moreover, we developed deep forecasting model that can predict the future prices of crude oil using our financial temporal patterns. The proposed model is tested and verified on the crude oil dataset with Root Mean Square Error (RMSE) economic test. Root Mean Square Error (RMSE) is a standard way to measure the error of a model in predicting quantitative data.