Examinando por Materia "Aprendizaje de máquina"

Mostrando 1 - 10 de 10
  • No hay miniatura disponible
    Ítem
    Aprendizaje reforzado profundo para la administración de portafolios de renta fija
    (Universidad EAFIT, 2023) Mejía Estrada, David; Almonacid Hurtado, Paula María
    This paper applies deep reinforced learning techniques to the management of fixed income investment portfolios, specifically sovereign securities issued by the Colombian government. The period of analysis covers seven years, from January 2015 to December 2022. We find that it is possible to generate profitability and achieve efficient risk management because of the trading strategies that deep reinforced learning models foresee more convenient given certain market conditions and of each of the securities, such as their implied risk in metrics like DV01, Duration and Convexity. Finally, this study contributes to the field of machine learning and artificial intelligence applications on investment portfolio management, with a relatively new focus on the fixed income market in general, consolidating itself as one of the first works to apply reinforcement learning techniques to the Colombian public debt market.
  • No hay miniatura disponible
    Ítem
    Estudio de la relación entre los valores sociales y la aceptación de sobornos como conducta corrupta : un estudio con modelos SEM y datos de la encuesta mundial de valores
    (Universidad EAFIT, 2024) Gómez Convers, Giovanny Hernando; Castrillón-Orrego, Sergio A.; Almonacid Hurtado, Paula María
    In a global context of rapid social change, investigating the relationship between social values and corruption has become increasingly urgent and significant. Which behaviors are desirable? Which do we manifest in daily life? The World Values Survey (WVS) serves as a crucial data source for understanding social values in various contexts. However, how these values influence the acceptance of bribery, and thus corruption, has not been sufficiently explored. This study examines the underlying patterns in response clusters and systematically analyzes them using the holistic possibilities offered by the institutionalism theoretical framework. The objective is to identify the most significant causalities and influences in the relationship between social values and corruption. Through robust data analysis, imputation techniques, dimensionality reduction, clustering analysis, and SEM modeling, we identify the main factors impacting the acceptance of bribery. The results demonstrate that the three pillars of institutionalism provide a valuable approach to understanding corruption by simultaneously considering key variables and components. When internalized, social values facilitate the acceptance of bribery in certain contexts, highlighting the influence of the cognitive dimension. Although legal frameworks can enhance transparency, cultural environment and customs have a more determining influence on the acceptance of corrupt practices. These findings underscore the need to foster a strong ethical culture and implement educational programs that promote integrity and transparency to effectively mitigate corruption.
  • No hay miniatura disponible
    Ítem
    Genomic Prediction and Genome-Wide Association Analysis in Common Bean (Phaseolus vulgaris l.) × Tepary bean (P. acutifolius a. gray) Inter-specific Advanced Lines at the Caribbean Coast of Colombia
    (Universidad EAFIT, 2023) López Hernández, Luis Felipe; Villanueva Mejía, Diego Fernando; Cortés Vera, Andrés Javier
    The negative effects of the climate change are risking global food security with 828 million people facing hunger, which is almost 16 times the population of Colombia. Given this scenario, legumes as common bean has offered a nature-based solution to source nutrients for rural communities in Latin America thanks to their high content of nutrients. For this reason, it is imperative to speed up the molecular genetic breading of common beans so that they can be cultivated in regions affected by extreme climate change, one of which is coastal Colombian. Therefore, in order to bridge this gap, this study aimed coupling an advanced panel of common bean (Phaseolus vulgaris L.) × tepary bean (P. acutifolius A. Gray) inter-specific lines with Bayesian regression algorithms to identify novel sources of adaptation to the humid and dry sub-regions at the Caribbean coast of Colombia, where common bean typically exhibits maladaptation to extreme heat waves. A total of 87 advanced lines with inter-specific ancestries were genotyped by sequencing (GBS), leading to the discovery of 15,645 single-nucleotide polymorphism (SNP) markers. Three yield components and two biomass variables were recorded for each genotype and inputted in several Bayesian regression models to identify the top genotypes with the best genetic breeding values across three localities in coastal Colombia. We explored the comparative analysis of several regression approaches where the model with the best performance in all traits and environments was BayesC. Also, we compared the utilization of all markers and only those determined as associated by a priori GWAS models. Better prediction ability with the complete SNP set was indicative of missing heritability as part of GWAS reconstructions. Furthermore, optimal SNP sets per environment and trait were determined to the top 500 most explicative markers according to their β regression effects. These 500 SNPs on average overlapped in 5.24 % across localities, which reinforced the environmentally dependent nature of polygenic adaptation. Finally, we retrieved the genomic estimated breeding values (GEBVs), and selected the top 10 genotypes for each environment and trait as part of a recommendation scheme targeting narrow adaption. The genotypes and SNP markers identified in this study as candidates for abiotic stress have the potential to be used in the following cycles as part of the long-term bean breeding program for coastal tropical regions.
  • No hay miniatura disponible
    Ítem
    Hurto a personas en la ciudad de Medellín : análisis predictivo de la cantidad de casos en diferentes zonas de la ciudad a partir de modelos de machine learning implementando técnicas de MLOps
    (Universidad EAFIT, 2023) Arboleda Colorado, Jeferson Stiven; Martínez Vargas, Juan David
    Robbery of individuals in Medellín is an issue demanding immediate attention. This prompted the study of the phenomenon within an analytics project, spanning data collection, database construction, modeling, and production deployment. It's worth noting that MLOps methodology was employed utilizing AWS services. Visual tools related to the phenomenon were integrated, facilitating analysis.
  • No hay miniatura disponible
    Ítem
  • No hay miniatura disponible
    Ítem
  • No hay miniatura disponible
    Ítem
    Optimización de portafolios financieros mediante enfoques de machine learning y computación cuántica : un caso de estudio
    (Universidad EAFIT, 2024) Agudelo Zuluaga, Mariana; Almonacid Hurtado, Paula María; Lalinde Pulido, Juan Guillermo
  • No hay miniatura disponible
    Ítem
    Predicción del cargue de rutas de distribución mediante aprendizaje de máquina
    (Universidad EAFIT, 2023) Ramírez Aguilar, Santiago; Téllez Falla, Diego Fernando; Marentes Cubillos, Luis Andrés
  • No hay miniatura disponible
    Ítem
    Stock Market Forecasting : Comparing Machine Learning and Deep Learning with Risk-Return Model Selection and Evaluation in a Walk-forward Approach
    (Universidad EAFIT, 2024) Castro Marín, Carlos Andrés; Olarte Hernández, Tomás; Olarte Hernández, Tomás
    This study compares the effectiveness of machine learning and deep learning algorithms in forecasting stock market direction using daily market data of Apple Inc. stock. We aim to determine if these algorithms can identify repeatable patterns across time using price and volume history and assess which are most capable. To ensure robustness, we employ a walk-forward validation approach to maintain the temporal dimension of the data and simulate real trading conditions. This method allows us to test models across different market conditions and measure their predictive power. We prioritize model selection and evaluation based on financial return and risk metrics, focusing on profitability rather than traditional machine learning performance metrics, which often do not correlate with financial outcomes. Our findings show that traditional machine learning algorithms, specifically Random Forest, outperform deep learning models under the selected asset and conditions tested. Machine learning models exceed the stock benchmark regarding Sharpe ratio, while deep learning models struggle to manage risk effectively, leading to poorer performance. This discrepancy is likely due to the complex solution space deep learning algorithms navigate to optimize and the amount of data required by these models. However, we hypothesize that the latter could improve its performance if tested with different architectures and hyperparameters, including newly developed transformer attention-based architectures and models such as TimeGPT and others, shown in the related work section.
  • No hay miniatura disponible
    Ítem
    Valoración del riesgo de crédito de empresas aplicando métodos analíticos e inteligencia artificial
    (Universidad EAFIT, 2023) Montoya Arias, José Andrés; Támara Ayús, Armando Lenin