Publicação

Financial distress in european vineyards and olive groves

Ver documento

Detalhes bibliográficos
Resumo:This study focuses on the prediction of financial distress of agricultural firms operating in the vineyards and olive crops sectors in Mediterranean countries, specifically in Portugal, Spain,and Italy, which are considered to be crucial for the production of these crops. The sample size of the study is 5,057 firms. Twelve models are presented, estimated from subsamples of combinations between countries and crops. Logistic regression is used for the estimation of these models.The accuracy of the models is evaluated, taking into account the importance of misclassification costs.Additionally, the areas under the ROC curves are calculated and compared in a dynamic of possible combinations between crops and countries. The study concludes that there are differences between the two sectors, as well as across countries, and suggests that dedicated models for each country or crop may improve the models’accuracy.
Autores principais:Céu, Mário S.
Outros Autores:Gaspar, Raquel M.
Assunto:Agriculture Financial Distress Prediction Models ROC Curves
Ano:2023
País:Portugal
Tipo de documento:working paper
Tipo de acesso:acesso aberto
Instituição associada:Universidade de Lisboa
Idioma:inglês
Origem:Repositório da Universidade de Lisboa
Descrição
Resumo:This study focuses on the prediction of financial distress of agricultural firms operating in the vineyards and olive crops sectors in Mediterranean countries, specifically in Portugal, Spain,and Italy, which are considered to be crucial for the production of these crops. The sample size of the study is 5,057 firms. Twelve models are presented, estimated from subsamples of combinations between countries and crops. Logistic regression is used for the estimation of these models.The accuracy of the models is evaluated, taking into account the importance of misclassification costs.Additionally, the areas under the ROC curves are calculated and compared in a dynamic of possible combinations between crops and countries. The study concludes that there are differences between the two sectors, as well as across countries, and suggests that dedicated models for each country or crop may improve the models’accuracy.