The present article presents a model LSTM for the forecast of product sales, an alternative in deep learning for this type of dilemmas and not frequent in the area of financial knowledge. It was approached as a time series and following the steps for the construction of models with machine learning. The ILE company of Ecuador provided the data, between 2011 and 2018. The results showed this model has a minimum RMSE error of 2.20 compared to another two models: ARIMA and Single Perceptron.
CITATION STYLE
Ordoñez-Ordoñez, P. F., Suntaxi Sarango, M. C., Narváez, C., Ruilova Sánchez, M. del C., & Cueva-Hurtado, M. E. (2020). Deep Learning Model for Forecasting Financial Sales Based on Long Short-Term Memory Networks. In Advances in Intelligent Systems and Computing (Vol. 1066, pp. 498–506). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-32022-5_46
Mendeley helps you to discover research relevant for your work.