A forecasting method of pharmaceutical sales based on ARIMA-LSTM model

Abstract

In order to improve the accuracy of sales forecasting for manufacturing companies, a combined forecasting method based on ARIMA model and long and short term memory (LSTM) neural network is proposed. ARIMA and LSTM are used to process the linear and nonlinear components of the sales time series, and then the prediction results are trained by the second LSTM model to obtain the final prediction. The results show that the combined model has a higher predictive precision and a wider range of applications.