Examinando por Materia "Forecasting"
Mostrando 1 - 20 de 26
Resultados por página
Opciones de ordenación
Ítem Affine term structure models: forecasting the Colombian yield curve(Universidad EAFIT, 2015-12-02) Velásquez-Giraldo, Mateo; Restrepo-Tobón, Diego A.; mvelas26@eafit.edu.co; drestr16@eafit.edu.coSuperior modeling of the yield curve is useful for asset pricing, financial planning, and risk management. In this article, we estimate five affine term structure models using daily Colombian data. We find that a three-factor model outperforms the other models in one and five days ahead forecasts. The model’s factors closely mimic empirical proxies for the level, the slope, and the curvature of the Colombian yield curve.Ítem AFOR-TSM: A Tool for Forecast with Time Series Models(Inst of Industrial Engineers; Cdr edition, 2012-01-01) Castro Zuluaga; Builes; Rosana; Bravo; Maria Cristina; Universidad EAFIT. Departamento de Ingeniería de Producción; Gestión de Producción y LogísticaOne of the most complex and necessary subjects in courses of Operations Management (OM) is that students understand and apply Time Series Models (TSM) to forecast future demand.Ítem Análisis y predicción de ventas de motos haciendo uso de la metodología “Customer Value Map” y técnicas de Machine Learning(Universidad EAFIT, 2024) Díaz Cordero, Sandra Marcela; Martínez Vargas, Juan David; Vallejo Correa, Paola AndreaÍtem An approach to make statistical forecasting of products with stationary/seasonal patterns(Production and Operations Management Society (POMS), 2014-05-09) Castro-Zuluaga, Carlos A.; Botero-Escobar, Sara C.; Castro-Zuluaga, Carlos A.; Botero-Escobar, Sara C.; Universidad EAFIT. Grupo de Investigación Gestión de Producción y Logística; Universidad EAFIT. Departamento de Ingeniería de Producción; ccastro@eafit.edu.co; sboter11@eafit.edu.co; Gestión de Producción y LogísticaAt any company there are hundreds, maybe thousands of products that must be forecasted with the best accuracy, in order to make a good demand management process, which are required at all planning levels in a company -- Statistical forecasting must be done fast and sometimes there are not enough resources to do it well -- In In this paper we make a proposal of an approach to define the parameter of the exponential smoothing model for products with a behavior pattern stationary or seasonal/stationary in the historical data, to obtain “good forecasting" A numerical example is used to show the effectiveness of the proposed methodÍtem Are neural networks able to forecast nonlinear time series with moving average components?(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 2015-07-01) Cogollo, M.R.; Velásquez, J.D.; Universidad EAFIT. Escuela de Ciencias; Modelado MatemáticoIn nonlinear time series forecasting, neural networks are interpreted as a nonlinear autoregressive models because they take as inputs the previous values of the time series. However, the use of neural networks to forecast nonlinear time series with moving components is an issue usually omitted in the literature. In this article, we investigate the use of traditional neural networks for forecasting nonlinear time series with moving average components and we demonstrate the necessity of formulating new neural networks to adequately forecast this class of time series. Experimentally we show that traditional neural networks are not able to capture all the behavior of nonlinear time series with moving average components, which leads them to have a low capacity of forecast. © 2015 IEEE.Ítem ¿Cambiará el paradigma? la minería de datos en la ciencia económica(Universidad EAFIT, 2017) Florez Llano, Andrés; N/AÍtem Desarrollo y aplicación de modelo neuroborroso para la predicción del precio de la acción de Ecopetrol en la Bolsa de Valores de Colombia mediante la utilización de indicadores técnicos y fundamentales y variables macroeconómicas(Universidad EAFIT, 2015) Rendón Gómez, Camilo; Mondragón Trujillo, Luis FernandoLos modelos de predicción de precios de activos financieros han captado la atención de investigadores e inversionistas debido a las posibles rentabilidades que se podrían obtener en caso de estimar valores acertados -- El presente trabajo presenta un modelo neuroborroso en el cual se combinan indicadores técnicos y fundamentales y variables macroeconómicas, comprendidas entre enero de 2010 y febrero de 2015, para predecir el precio de la acción de Ecopetrol en la Bolsa de Valores de Colombia -- Algunas variables macroeconómicas se modelan mediante sistemas de inferencia borrosos y sus salidas se agregan a los indicadores técnicos y fundamentales para entrar a una red neuronal artificial de tipo anticipativo (feedforward) que pronostica el precio del activo para el mes siguienteÍtem Diseño de un modelo para la proyección, seguimiento y control del flujo de efectivo en las pymes del sector manufacturero, comercializador y servicios. Caso aplicado a los clientes de Covalentte SAS(Universidad EAFIT, 2024) España García, Daniela; Torres Sepúlveda, Carlina Yamile; Orozco Echeverry, César AugustoThe main objective of this field work was to develop a model that allows building projections, monitoring and control of cash flow in SMEs in the manufacturing, trading and services sectors, being tested in companies that are currently clients of Covalentte SAS, a company dedicated to providing accounting, tax, and legal advisory services to organizations with different business activities. To develop the model, the financial statements, accounting information, indicators and cash flow structure of the companies under study were analysed, in order to incorporate quantitative forecasting methods based on time series to predict future results in the short and medium term, as well as simulation for decision making. Finally, it was possible to test the model in 15 companies from different sectors which, through the execution of a survey, allowed validating the level of acceptance of the process, being the service companies those who gave it a better valuation because their sales, which in most cases are cash and do not incorporate the inventory component in their working capital structure, simplify with it their cash flow management, allowing informed decition maling by the companies managers.Ítem Do U.S. and Colombian macro factors improve the forecasting ability of unrestricted VAR models of the local term structure of interest rate?(Universidad EAFIT, 2015) Arteaga Estrada, Laura; Restrepo Tobón, Diego AlexanderEste estudio empírico compara la capacidad de los modelos Vectores auto-regresivos (VAR) sin restricciones para predecir la estructura temporal de las tasas de interés en Colombia -- Se comparan modelos VAR simples con modelos VAR aumentados con factores macroeconómicos y financieros colombianos y estadounidenses -- Encontramos que la inclusión de la información de los precios del petróleo, el riesgo de crédito de Colombia y un indicador internacional de la aversión al riesgo mejora la capacidad de predicción fuera de la muestra de los modelos VAR sin restricciones para vencimientos de corto plazo con frecuencia mensual -- Para vencimientos de mediano y largo plazo los modelos sin variables macroeconómicas presentan mejores pronósticos sugiriendo que las curvas de rendimiento de mediano y largo plazo ya incluyen toda la información significativa para pronosticarlos -- Este hallazgo tiene implicaciones importantes para los administradores de portafolios, participantes del mercado y responsables de las políticasÍtem Dynamic cost forecasting Drayage Product in inland transport using machine learning models(Universidad EAFIT, 2023) Peralta Jaramillo, Juliana Andrea; Moreno Reyes, Nicolás AlbertoÍtem Ensemble of temporal convolutional and long short-term memory neural networks apply to forecasting USDCOP exchange rate(Universidad EAFIT, 2021) Torres Marulanda, Juan Esteban; Almonacid Hurtado, Paula MaríaThis paper applies a neural network with ensemble of temporal convolutional network (TCN) and long short-term memory (LSTM) layers approach to forecast foreign exchange rates between the US dollar (USD) and Colombian Peso (COP) and obtain a better performance. This study provides evidence on the TCN and LSTM neural network model’s effectiveness and efficiency in forecasting temporal series. It should contribute positively to developing theory, methodology, and practice of using an artificial neural network to develop a forecasting model for financial temporal series.Ítem FOR-TSM: desarrollo de una herramienta de pronósticos con modelos de series de tiempo(Universidad EAFIT, 2011) Bravo Gómez, María Cristina; Builes Palacio, Rosana; Castro Zuluaga, Carlos AlbertoUno de los temas más complejos y necesarios en los cursos de Administración de Operaciones, es el uso de los pronósticos con modelos de series de tiempo (TSM por sus siglas en inglés) -- Para facilitar el entendimiento y ayudar a los estudiantes a comprender fácilmente los pronósticos de demanda, este proyecto presenta FOR TSM, una herramienta desarrollada en MS Excel VBA® -- La herramienta fue diseñada con una Interfaz gráfica de Usuario (GUI por sus siglas en inglés) para explicar conceptos fundamentales como la selección de los parámetros, los valores de inicialización, cálculo y análisis de medidas de desempeño y finalmente la selección de modelosÍtem Improvement of a knock model for natural gas SI engines through heat transfer evaluation(Springer-Verlag France, 2018-11-01) Sierra Parra A.F.; Díaz Torres A.G.; Sierra Parra A.F.; Díaz Torres A.G.; Universidad EAFIT. Departamento de Ingeniería de Producción; Ingeniería, Energía, Exergía y Sostenibilidad (IEXS)Knock is an abnormal combustion phenomena capable of causing serious damage to spark ignition engines, and is a constraint to reach the maximum potential of the engine, since strategies to increase power output and improve efficiency such as turbocharging, increased compression ratio and the advancement of spark timing, also increase the possibility of knock occurrence. Therefore, it is crucial to take into account the limits imposed by knock in the design and operating conditions of the engine when using an engine computational model. In this article a zero-dimensional two-zone engine model, coupled with a chemical kinetic model for knock detection through end-gas auto-ignition is developed and validated, for a natural gas engine. Given the importance of an accurate knock prediction, five heat transfer coefficient correlations are compared to find the most suitable to predict the knock occurrence, through calculation of a knock criterion. Correlations from Sitkei and Annand were the most suitable to predict this knock criterion for the experimental data used, and the Sitkei correlation was later tested in a parametric study to predict the effect of spark timing, compression ratio, equivalence ratio and inlet temperature in knock occurrence and intensity. Results were in accordance with real engine behaviour when knock occurs. © 2017, Springer-Verlag France SAS, part of Springer Nature.Ítem Methodological advances in artificial neural networks for time series forecasting(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 2014-06-01) Cogollo, M. R.; Velasquez, J. D.; Universidad EAFIT. Escuela de Ciencias; Modelado MatemáticoObjective: The aim of this paper is to analyze the development of new forecasting models based on neural networks. Method: We used the systematic literature review method employing a manual search of papers published on new neural networks models in the time period 2000 to 2010. Results: Only 18 studies meet all the requirements of the inclusion criteria. Of these, only three proposals considered a neural networks model using a process different to the autoregressive. Conclusion: Although studies relating to the application of neural network models were frequently present, we find that the studies proposing new forecasting models based on neural networks with a theoretical support and a systematic procedure for the construction of model, were scarce in the time period 2000-2010. © 2012 IEEE.Ítem Modelamiento predictivo del número de visitantes en un centro comercial(Universidad EAFIT, 2022) Rua Jaramillo, Ramón David; Laniado Rodas, Henry; Almonacid Hurtado, Paula MaríaThe ability to make predictions about the number of customers or visitors in a shopping center is a very important input in the planning and efficient use of physical and human resources in this type of company. Also, it is important to understand what aspects influences their behavior. Based on historical data on the number of visitors, as well as external (environment) variables and online search trends, a forecasting model of the behavior of daily visits to the shopping center is suggested. The historical data correspond to the pedestrian and vehicular entries (cars and motorcycles) of the last 6 years in a shopping center located in the city of Medellín. This project begins with a literature review regarding forecasting models in different places such as museums, airports, natural parks, shopping centers and restaurants, among others, in order to explore methodologies in such cases and possible solution options. Through time series analysis and machine learning algorithms, the most representative variables and the best-fit model are selected to predict the number of visitors. This model is expected to be strengthened with estimation algorithms, improving performance over time and allowing it to be applied in other business or educational environments.Ítem Modelo de predicción de insolvencia financiera aplicado al sector farmacéutico colombiano(Universidad EAFIT, 2015) Penagos Girón, José Luis; Muñoz Herrera, Óscar AndrésEl presente trabajo presenta el desarrollo de un modelo Logit para datos de panel desbalanceado que permite calcular la probabilidad que las empresas del sector farmacéutico en Colombia incurran en insolvencia financiera -- Se toma el modelo Logit para lograr predecir la probabilidad de una forma temprana una quiebra empresarial -- Para construir el modelo se toma la información de los estados financieros de las empresas que pertenecen al sector farmacéutico de las bases de datos de la Superintendencia de Sociedades de Colombia durante el período 2008-2013 -- Para la creación de nuestro modelo se tendrá en cuenta la revisión, análisis y comparación de los distintos modelos de predicción de bancarrota empresarial trabajados a nivel nacional e internacional, en los que se consideran tanto aspectos macroeconómicos, como las crisis financieras mundiales, las dinámicas económicas internacionales, reformas tributarias de un país, conflictos sociales y tasas de interés; como los microeconómicos, tales como los desempeños operacionales y los financieros empresariales -- La suma de estos factores internos y externos influyen drásticamente en el funcionamiento de una empresa -- A partir de la información financiera de las empresas del sector farmacéutico se establecen distintos ratios de desempeño financiero y operacional, que se involucran en el modelo econométrico Logit, para predecir la probabilidad de insolvencia empresarial de una forma temprana, aspecto que podrá ser considerado en la gestión la quiebra d integral de las mismasÍtem Predicción de precios de materias primas por medio del seguimiento de indicadores económicos(Universidad EAFIT, 2015) Gómez Gómez, Daniel; Bedoya García, Cristian Camilo; Cuéllar Bermúdez, Ulises OrestesLa predicción del comportamiento de la economía mundial se ha hecho más compleja con la globalización de las últimas décadas -- Al tener acceso a información en vivo, todos los fenómenos económicos se transmiten fácilmente de un lado del planeta al otro, afectando los mercados e industrias a un mayor ritmo que anteriormente -- Además de esto, esta sinergia ha llevado a una alta volatilidad de precios de las materias primas -- Sin embargo, existen métodos que permiten hacer un acercamiento a esta predicción -- En el Grupo Familia, la compra de materias primas constituye un porcentaje alto de la torta de participación de compras totales -- Debido al alto costo que representan estas compras para el Grupo y a las constantes fluctuaciones en la economía mundial, se ha detectado una oportunidad de mejora y ahorro si se logra predecir el comportamiento de los precios de dichas materias primas -- En este trabajo se evaluarán las probabilidades de poder predecir las fluctuaciones de precios por medio de un seguimiento a indicadores económicos -- La investigación definirá qué materias primas y cuáles indicadores económicos serán sujeto de prueba -- Al final, se pretende poder entregar los resultados que le permitan al Grupo Familia poder predecir el comportamiento de las materias primas para tomar decisiones acertadas y tener una posición privilegiada de negociación ante los proveedoresÍtem Predicción del cargue de rutas de distribución mediante aprendizaje de máquina(Universidad EAFIT, 2023) Ramírez Aguilar, Santiago; Téllez Falla, Diego Fernando; Marentes Cubillos, Luis AndrésÍtem Prediction of a flying droplet landing over a non-flat substrates for ink-jet applications(Springer-Verlag France, 2019-01-01) Arango I.; Bonil L.; Posada D.; Arcila J.; Arango I.; Bonil L.; Posada D.; Arcila J.; Universidad EAFIT. Departamento de Ingeniería Mecánica; Mecatrónica y Diseño de MáquinasPrinting with inkjet technology has found new forms of application in the industry and in this article we study this technology focused on printing on non-flat surfaces. Since there is no print history over distances greater than 1 mm due to the rupture phenomenon, an initial quality standard is defined to measure achievements in a relative manner. An interactive method is used that requires the user to approach the machine in multiple analyzes of different types. The first approach is a mathematical model this model was constructed to predict the drop distance of the drop in the non-planar substrate with respect to the planned one in the flat substrate, taking into account that most of the drops fall to different heights presenting a greater or lesser state of development the phenomena present in the flight. The results allow to initiate a process of compensation that avoids the distortion of the figure to improve the printing resolution. The results are validated using a relative quality through industrial ink-jet printer with heads capable of injecting functional fluids. The initial result indicates that in standard surface printing with print relative quality already defined, it can be used only for low resolution formats with thick lines, and the result can be improved when the original figure is treated by compensating the distance between the numerical prediction and the initial objective. © 2019, Springer-Verlag France SAS, part of Springer Nature.Ítem Prediction of Federal Funds Target Rate: a dynamic logistic Bayesian Model averaging approach(Universidad EAFIT, 2015) Alzate Arias, Hernán Alonso; Ramírez Hassan, AndrésIn this paper we examine which macroeconomic and financial variables have most predictive power for the target repo rate decisions made by the Federal Reserve -- We conduct the analysis for the FOMC decisions during the period June 1998-April 2015 using dynamic logistic models with dynamic Bayesian Model Averaging that allows to perform predictions in real-time with great flexibility -- The computational burden of the algorithm is reduced by adapting a Markov Chain Monte Carlo Model Composition: MC3 -- We found that the outcome of the FOMC meetings during the sample period are predicted well: Logistic DMA-Up and Dynamic Logit-Up models present hit ratios of 87,2 and 88,7; meanwhile, hit ratios for the Logistic DMA-Down and Dynamic Logit-Down models are 79,8 and 68,0, respectively