A carregar...
Publicação
Combinação da lei de benford e de um modelo baseado nos accruals discricionários para aferir a qualidade da informação financeira
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.18 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
A presente dissertação tem como objetivo avaliar a eficácia da combinação de duas proxies utilizadas na literatura para aferir a Qualidade da Informação Financeira (QIF): Lei de Benford e um modelo baseado nos accruals discricionários. A combinação destas duas formas de aferir a QIF pode fornecer uma metodologia mais robusta e integrada para aferir a QIF e identificar potenciais manipulações nos relatórios financeiros das empresas. A Lei de Benford é uma técnica estatística utilizada para detetar anomalias numéricas em grandes conjuntos de dados, enquanto que o modelo de accruals discricionários, nomeadamente o Modelo de Jones, utiliza os accruals para identificar manipulações de resultados. A investigação classificou, numa primeira fase, as empresas do setor da restauração e similares com indícios de manipulação dos resultados, por um lado, com recurso à aplicação da Lei de Benford e, por outro lado, com recurso à aplicação do modelo de Jones. Numa segunda fase, pretendeu-se identificar as empresas com indícios de manipulação dos resultados que sejam classificadas em simultâneo por ambos os métodos. Em cada uma das fases relacionou-se a classificação das empresas com indícios de manipulação de resultados com o incentivo de minimização do pagamento de impostos ao Estado. A metodologia inclui uma análise detalhada dos resultados obtidos pela aplicação de cada técnica individualmente e em conjunto. Em suma, os resultados indicam que a combinação dos dois métodos melhora de forma substancial a capacidade de deteção de manipulação e evidencia indícios localizados de minimização fiscal, demonstrando o potencial destas metodologias como instrumentos complementares de auditoria e controlo fiscal.
The present dissertation aims to evaluate the effectiveness of combining two proxies used in the literature to measure Financial Information Quality (QIF): Benford's Law and a model based on discretionary accruals. The combination of these two ways of measuring QIF provided a more robust and integrated methodology for measuring QIF and identifying potential manipulations in companies' financial reports. Benford's Law is a statistical technique used to detect numerical anomalies in large data sets, while the discretionary accruals model, namely the Jones Model, uses accruals to identify earnings manipulation. The research initially classifies companies in the restaurant and related sectors with signs of earnings manipulation, through the application of Benford’s Law on the one hand, and through the application of the Jones model on the other. In a second phase, it was intended to identify companies with signs of earnings manipulation that are classified simultaneously by both methods. In each of the phases, the classification of companies with evidence of earnings manipulation was related to the incentive to minimise tax payments to the State. The methodology includes a detailed analysis of the results obtained by applying each technique individually and in combination. In conclusion, the results indicate that combining the two methods substantially enhances the ability to detect manipulation and provides localised evidence of tax minimisation, demonstrating the potential of these methodologies as complementary instruments for auditing and tax control.
The present dissertation aims to evaluate the effectiveness of combining two proxies used in the literature to measure Financial Information Quality (QIF): Benford's Law and a model based on discretionary accruals. The combination of these two ways of measuring QIF provided a more robust and integrated methodology for measuring QIF and identifying potential manipulations in companies' financial reports. Benford's Law is a statistical technique used to detect numerical anomalies in large data sets, while the discretionary accruals model, namely the Jones Model, uses accruals to identify earnings manipulation. The research initially classifies companies in the restaurant and related sectors with signs of earnings manipulation, through the application of Benford’s Law on the one hand, and through the application of the Jones model on the other. In a second phase, it was intended to identify companies with signs of earnings manipulation that are classified simultaneously by both methods. In each of the phases, the classification of companies with evidence of earnings manipulation was related to the incentive to minimise tax payments to the State. The methodology includes a detailed analysis of the results obtained by applying each technique individually and in combination. In conclusion, the results indicate that combining the two methods substantially enhances the ability to detect manipulation and provides localised evidence of tax minimisation, demonstrating the potential of these methodologies as complementary instruments for auditing and tax control.
Descrição
Palavras-chave
Lei de Benford Modelo de accruals discricionários Anomalias Manipulações de resultados