Search academic texts

Information × Registration Number 2123U006681, Article popup.category Препринт Title popup.author Bilinska Marta popup.publication 01-01-2023 popup.source_user Український католицький університет popup.source https://hdl.handle.net/20.500.14570/4812 popup.publisher Description Demand and revenue forecasting are crucial for a lot of business processes which include resource planning, financial management, and optimization of pricing strategies. This work explores the approach to forecasting demand in a competitive market by applying various time series forecasting methods. It focuses on developing accurate and flexible models to predict the future demand for products of a company which operates on the Ukrainian market. It also explores the specifics of demand forecasting in the competitive environment. The research covers a range of models for demand predictions, such as exponential smoothing, SARIMAX, ARDL, machine learning methods, and neural networks like LSTM. Furthermore, the study also examines the impact of exogenous variables and feature engineering on different model performances. The research proposes the bestsuiting approach to demand and revenue forecasting for the examined company, while the power of the model is evaluated using various performance metrics such as MAPE and RMSE. Additionally, the work provides forecasts for the company’s demand and revenue for the period of 2023–2024 and presents recommendations based on those forecasts. popup.nrat_date 2025-05-09 Close