Resumen de Análise de Fundos Imobiliários com Redes Neurais e Neural Tools

Mardoqueu Josue Rufino, Rosemere da Silva Florencio, Victor de Souza Barbosa, Alexandre Leme Sanches

• español

  Resumen Este estudio utiliza redes neuronales artificiales (RNAs) para predecir el desempeño de los Fondos de Inversión Inmobiliaria (FIIs). Se analizaron seis FIIs con base en datos de Funds Explorer, usando los indicadores de liquidez diaria, P/Valor Libro (P/VL) y Rendimiento por Dividendos (DY) acumulado de 12 meses. Los resultados indican que las RNAs pueden predecir con precisión el desempeño de los FIIs, ofreciendo un enfoque innovador para inversores y gestores.

  Palabras clave: Fondos de Inversión Inmobiliaria; Redes Neuronales Artificiales; Predicción; Análisis de Datos; Inversiones.

• português

  R E S U M O Este estudo utiliza redes neurais artificiais (RNAs) para prever o desempenho de Fundos de Investimentos Imobiliários (FIIs). Seis FIIs foram analisados com base em dados do Funds Explorer, usando os indicadores de liquidez diária, P/VP e DY (12M) acumulado. Os resultados indicam que as RNAs podem prever o desempenho dos FIIs com precisão, oferecendo uma abordagem inovadora para investidores e gestores.

  Palavras-chave: Fundos de Investimento Imobiliário; Redes Neurais Artificiais; Previsão; Análise de Dados; Investimentos.

• English

  Objective: The study analyzed the application of artificial neural networks (ANNs) in predicting Real Estate Investment Funds (REITs) in Brazil, using data such as daily liquidity, P/NAV, and accumulated Dividend Yield (DY) to identify patterns and trends.

  Theoretical Framework: The theoretical framework discusses the evolution of REITs in Brazil since the 1990s, highlighting their growing popularity among average investors and the complexity in data analysis. Authors like Mueller and Mueller (2003) emphasize the importance of statistical and fundamental methods in evaluating REITs.

  Method: The study used NeuralTools software to process data collected from REITs, focusing on indicators such as daily liquidity, P/NAV, and accumulated DY (12M), with the aim of predicting the future performance of the funds.

  Results and Discussion: The results showed a high accuracy rate in the predictions made by the ANNs, indicating a significant relationship between the analyzed indicators and the performance of the REITs. However, the research also pointed out the need for further studies to fully explore the potential of ANNs.

  Research Implications: The use of ANNs can enhance investment decision-making in REITs, providing valuable insights. The recommendations may serve as a basis for implementing artificial intelligence technologies in the financial sector.

  Originality/Value: The study contributes to the literature on REIT investments, presenting an innovative approach to the analysis and management of real estate investments in Brazil, as well as suggesting the adoption of artificial intelligence techniques.