Search academic texts

Information × Registration Number 2123U006685, Article popup.category Препринт Title popup.author Kmet Diana-Sofiia popup.publication 01-01-2023 popup.source_user Український католицький університет popup.source https://hdl.handle.net/20.500.14570/4809 popup.publisher Description Financial planning has always been a pressing issue for businesses. In order to stay competitive in today’s fast-paced environment, many companies apply forecasting to estimate their future sales. This study addresses the problem of a small chain of grocery stores in Western Ukraine, which has faced significant challenges due to turbulent events such as the COVID-19 pandemic and ongoing war. This work focuses on predicting sales for the mentioned grocery retailer by employing a range of forecasting techniques, includ- ing SARIMA, Simple Exponential Smoothing, Holt’s linear Exponential Smoothing, XGboost, RNN, and LSTM. The objective is to identify the best-performing model given the challenges posed by a volatile dataset. In addition, this study aims to provide business insights on seasonal patterns in the sales data for all stores within the chain. Furthermore, it explores the impact of the war and the pandemic on sales and identifies other important features that affect sales forecasting performance. popup.nrat_date 2025-05-09 Close