2021/01/08 by Racine Ly, Ly, Racine, Fousseini Traoré +3 · 1 citation
Decision Sciences · Economics, Econometrics and Finance · Engineering · #Energy Load and Power Forecasting #FOS: Computer and information sciences #FOS: Economics and business #I.6.6 #Machine Learning (cs.LG) #Market Dynamics and Volatility #Statistical Finance (q-fin.ST) #Stock Market Forecasting Methods
paper · pdf · doi:10.48550/arxiv.2101.03087
openalex publication_date 2021/01/08 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
This paper applies a recurrent neural network (RNN) method to forecast cotton\nand oil prices. We show how these new tools from machine learning, particularly\nLong-Short Term Memory (LSTM) models, complement traditional methods. Our\nresults show that machine learning methods fit reasonably well the data but do\nnot outperform systematically classical methods such as Autoregressive\nIntegrated Moving Average (ARIMA) models in terms of out of sample forecasts.\nHowever, averaging the forecasts from the two type of models provide better\nresults compared to either method. Compared to the ARIMA and the LSTM, the Root\nMean Squared Error (RMSE) of the average forecast was 0.21 and 21.49 percent\nlower respectively for cotton. For oil, the forecast averaging does not provide\nimprovements in terms of RMSE. We suggest using a forecast averaging method and\nextending our analysis to a wide range of commodity prices.\n