FORECASTING DEMAND IN RETAIL INDUSTRY USING AUTOREGRESSIVE TECHNIQUES
No Thumbnail Available
Date
2024-04
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Nazarbayev University School of Sciences and Humanities
Abstract
This paper aims to compare the accuracy of different forecasting methods in retail industry in Ecuador. In particular, autoregression, autoregressive moving average, vector autoregression, and neural network autoregression were used to predict the demand for three time series. The study used the data provided by “Corporacion Favorita” from Kaggle. The three time series that were used are “automotive”, “dairy” and “beverages” which correspond to the product categories. The study finds that for “dairy” and “beverages” product families autoregressive model is the most accurate, while neural network autoregressive model is the most accurate for the “automotive” category. For the vector autoregressive model price of “Brent” oil was used, but this model was the least accurate. The accuracy of the autoregressive moving average model was in between autoregressive and vector autoregressive models.
Description
Keywords
Type of access: Gated
Citation
Bolat, A. (2024). Forecasting demand in retail industry using autoregressive techniques. Nazarbayev University School of Sciences and Humanities