Examinando por Materia "XGBoost"
Mostrando 1 - 7 de 7
Resultados por página
Opciones de ordenación
Publicación Análisis de riesgo de impago en el sector financiero : enfoque en tarjetas de crédito(Universidad EAFIT, 2024) Herrera Olivares, Maher Stehisy; Moreno Reyes, Nicolás AlbertoPublicación Aplicación de modelos predictivos tradicionales y de machine learning para evaluar la insolvencia en el sector salud de Colombia(Universidad EAFIT, 2025) Girón Ocampo, César Eduardo; Mercado Vargas, Luis Fernando; Cruz Castañeda, Vivian; Álvarez Franco, Pilar BeatrizPublicación Clasificación de créditos de libranza negociados en el mercado secundario colombiano, aplicando técnicas de aprendizaje supervisado(Universidad EAFIT, 2024) Gómez Betancur , Juan Camilo; Moreno Reyes, Nicolás AlbertoCredit risk, exacerbated by events such as the 2008 financial crisis, remains a concern for both banking and non-banking entities. This study addresses the need to improve the classification of payroll loans in Colombia using both traditional and machine learning techniques. It highlights the superior effectiveness of supervised learning algorithms in credit risk classification, with the ultimate goal of developing a model capable of identifying loans with a higher probability of default. This would optimize the acquisition of payroll loans and strengthen investment portfolio.Ítem Determinantes del riesgo de incumplimiento en créditos educativos : un análisis para Colombia(Universidad EAFIT, 2020) Granda Rodríguez, Manuela Andrea; Posso Suárez, Christian ManuelThis document uses non-parametric Machine Learning methodologies, in particular the XGBoost algorithm, to predict the risk of non-compliance with educational credits in Colombia offered by ICETEX between 2015 and 2018. The interest variable is the risk of default in student credits and is used as variable determinants associated with the socioeconomic level of students, as well as school information and academic achievement for each student. The main results show that socioeconomic variables with very good default predictors, in particular variables such as parent education and scores on critical reading tests are strong predictors. The results found contribute to economic and social policy decisions on the design of methods for higher education coverage through meritorious credits with public and private funds.Publicación Hacia un modelo predictivo para identificar aspectos clave en la rotación de empleados(Universidad EAFIT, 2023) Cárdenas López, Paula Andrea; Tabares Betancur, Marta Silvia; Postobon S.APublicación Modelo fundamental de crecimiento en utilidades y price-to-earnings ratio, P/E, de los índices accionarios internacionales(Universidad EAFIT, 2025) Jiménez Benítez, Daniel; Sandino Perdomo, Daniel; Navarrete Quintero, Nicolás; Diaz, Walter; Durango Gutiérrez, María PatriciaThe S&P500, a barometer of the U.S. economy, is one of the world's leading stock market indices. The price-to-earnings ratio (P/E) is a valuation measure that compares a stock's market price to its earnings per share, and is commonly used to assess whether stocks are overvalued or undervalued. Forecasting the P/E ratio is complex due to factors that can influence the ratio: interest rates, economic growth, market sentiment, and financial projections for companies, among others. In this research, two recurrent neural network models were implemented: LSTM (long short-term memory) and GRU (gated recurrent unit), as well as two machine learning models: XGBoost (extreme gradient boosting) and LigthGBM (light gradient boosting machine), to forecast the P/E ratio of the S&P500 using historical data between January 1990 and October 2024. The results show that all four models perform well, although the GRU model stands out in terms of accuracy and computational efficiency, without leaving aside the LightGBM model, a boosting algorithm, which also shows competitive results. The research offers valuable information on the use of the four models to forecast valuation ratios, and can be useful as support in investment decision making.Publicación Predictive Model for the Detection of Natural Gas Customers Who Consume the Service Without Being Billed(Universidad EAFIT, 2022) Jaramillo Zapata, Diego Andrés; Tabares Betancur, Marta Silvia; Tabares Betancur, Marta Silvia