Comparative study on retail sales forecasting between single and combination methods

Abstract

In today's competitive global economy, businesses must adjust themselves constantly to ever-changing markets. Therefore, predicting future events in the market-place is crucial to the maintenance of successful business activities. In this study, sales forecasts for a global furniture retailer operating in Turkey were made using state space models, ARIMA and ARFIMA models, neural networks, and Adaptive Network-based Fuzzy Inference System (ANFIS). Also, the forecasting performances of some widely used combining methods were evaluated by comparison with the weekly sales data for ten products. According to the best of our knowledge, this study is the first time that the recently developed state space models, also called ETS (Error-Trend-Seasonal) models, and the ANFIS model have been tested within combining methods for forecasting retail sales. Analysis of the results of the single models in isolation indicated that none of them outperformed all the others across all the time series investigated. However, the empirical results suggested that most of the combined forecasts examined could achieve statistically significant increases in forecasting accuracy compared with individual models and with the forecasts generated by the company's current system.

Keywords:

• sales forecasting
• neural networks
• exponential smoothing
• ARIMA
• ARFIMA
• ANFIS
• combined forecasts
• retail sales

JEL Classification:

• M21
• C53
• C45

Additional information

Notes on contributors

Serkan Aras

Serkan ARAS received his Bachelor degree in Econometrics in 2005 from Dokuz Eylul University, Izmir, Turkey and in 2008 obtained an MSc in Econometrics from Dokuz Eylul University. He got his PhD degree in Econometrics, in 2013 from Dokuz Eylul University. Also, he visited School of Computer Science, University of Birmingham, U.K. three semesters as academic researcher during his PhD. His research interests include neural networks, evolutionary algorithms, fuzzy logic and time series forecasting.

İpek Deveci Kocakoç

İpek DEVECİ KOCAKOÇ received her PhD in Econometrics with a major in Industrial Engineering from Dokuz Eylul University, Turkey. Her recent research interests include Artificial Intelligence, Scientific Programming and Quantitative methods. She is currently a full time professor in Dokuz Eylul University.

Cigdem Polat

Çiğdem POLAT received her Bachelor degree in Industrial Engineering in 2010 from Sakarya University, Turkey and in 2016 obtanied an MSc in Quality Management from Dokuz Eylul University. Her recent interests include Total Quality Management and Quantitative Methods.