Do Large Datasets or Hybrid Integrated Models Outperform Simple Ones in Predicting Commodity Prices and Foreign Exchange Rates?

With the continuous advancement of machine learning and increasing availability internet-based information, there is a belief that these approaches datasets enhance accuracy price prediction. However, this study aims to investigate validity claim. The examines effectiveness large dataset sophisticated methodologies in forecasting foreign exchange rates (FX) commodity prices. Specifically, we employ sentiment analysis construct robust index explore whether combining with surpasses performance when predicting FX Additionally, apply such as random forest (RF), eXtreme gradient boosting (XGB), long short-term memory (LSTM), alongside classical statistical model autoregressive integrated moving average (ARIMA), forecast prices compare models’ performance. Based on results, propose novel integrate wavelet transformation ARIMA techniques (seasonal-decomposition-ARIMA-LSTM, wavelet-ARIMA-LSTM, wavelet-ARIMA-RF, wavelet-ARIMA-XGB). We procedure gold futures euro against US dollar.