| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 206.6 KB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Este projeto, desenvolvido no âmbito do Mestrado em Matemática Aplicada à Economia e Gestão,
tem como objetivo melhorar a previsão de vendas de produtos no sector do retalho, com especial atenção
aos períodos em que ocorrem promoções. O estudo aborda a complexidade da previsão de vendas num
ambiente altamente competitivo, onde as promoções desempenham um papel crucial na estratégia de
marketing das empresas.
A investigação utiliza dados reais de uma empresa retalhista internacional (PromoPlus) para desen volver e avaliar modelos de regressão linear múltipla que incorporam variáveis promocionais. Estes
modelos são comparados com métodos mais simples, como o alisamento exponencial, para verificar
qual a abordagem que fornece previsões mais exactas.
Este projeto está dividido em cinco secções. A primeira contém a introdução e a revisão da literatura
e introduz os conceitos de previsão de vendas, promoção de vendas e gestão de stocks. Esta secção
também analisa criticamente as diferentes contribuições da literatura sobre este tema. A segunda secção
identifica as metodologias utilizadas para resolver a questão de investigação proposta. A terceira secção
apresenta a modelização para avaliar o impacto das promoções nas vendas. A quarta secção trata da
modelização e da previsão do volume de vendas. Por último, são sintetizados os resultados obtidos com
os ajustamentos dos modelos de regressão linear múltipla e com a previsão do volume de vendas.
Os resultados indicam que a inclusão de variáveis relacionadas com a promoção nos modelos de
previsão pode melhorar significativamente a precisão das previsões de vendas, ajudando os retalhistas a
otimizar a gestão das existências e a maximizar os lucros durante os períodos promocionais.
This project, carried out as part of the Master’s programme in Applied Mathematics for Economics and Management, aims to improve product sales forecasting in the retail sector, with special attention to periods when promotions take place. The study addresses the complexity of sales forecasting in a highly competitive environment, where promotions play a crucial role in companies’ marketing strategy. The research uses real data from an international retail company (PromoPlus) to develop and evaluate multiple linear regression models that incorporate promotional variables. These models are compared with simpler methods, such as exponential smoothing, to see which approach provides more accurate predictions. This project is divided into five sections. The first contains the introduction and literature review and introduces the concepts of sales forecasting, sales promotion and stock management. This section also critically analyses the different contributions in the literature on this topic. The second section identifies the methodologies used to solve the proposed research question. The third section presents the modelling to assess the impact of promotions on sales. The fourth section deals with modelling and sales volume forecasting. Finally, the results obtained with the adjustments to the multiple linear regression models and the sales volume forecast are summarised. The results indicate that including promotion-related variables in forecasting models can significantly improve the accuracy of sales forecasts, helping retailers to optimise stock management and maximise profits during promotional periods.
This project, carried out as part of the Master’s programme in Applied Mathematics for Economics and Management, aims to improve product sales forecasting in the retail sector, with special attention to periods when promotions take place. The study addresses the complexity of sales forecasting in a highly competitive environment, where promotions play a crucial role in companies’ marketing strategy. The research uses real data from an international retail company (PromoPlus) to develop and evaluate multiple linear regression models that incorporate promotional variables. These models are compared with simpler methods, such as exponential smoothing, to see which approach provides more accurate predictions. This project is divided into five sections. The first contains the introduction and literature review and introduces the concepts of sales forecasting, sales promotion and stock management. This section also critically analyses the different contributions in the literature on this topic. The second section identifies the methodologies used to solve the proposed research question. The third section presents the modelling to assess the impact of promotions on sales. The fourth section deals with modelling and sales volume forecasting. Finally, the results obtained with the adjustments to the multiple linear regression models and the sales volume forecast are summarised. The results indicate that including promotion-related variables in forecasting models can significantly improve the accuracy of sales forecasts, helping retailers to optimise stock management and maximise profits during promotional periods.
Descrição
Trabalho de Projeto de Mestrado, Matemática Aplicada à Economia e Gestão, 2024, Universidade de Lisboa, Faculdade de Ciências
Palavras-chave
Previsão de vendas Promoções Gestão de stocks Alisamento exponencial Regressão linear múltipla Teses de mestrado - 2024
