Examinando por Autor "Santana Contreras, Juan Camilo"
Mostrando 1 - 12 de 12
Resultados por página
Opciones de ordenación
- ÍtemAjuste de una cópula a un conjunto de datos longitudinales como alternativa a la construcción de la estructura de covarianza de un modelo de efectos fijosVillada Cantor, Diego Alexis; Santana Contreras, Juan Camilo; Modelo de efectos fijos - - Métodos Estadísticos
- ÍtemCambios aditivos e innovativos en series temporalesMendez Gamba, Elvin Rodrigo; Santana Contreras, Juan Camilo
- ÍtemContraste de pronósticos de ventas de plaguicidas para una empresa del sector agroquímico en la región Latinoamérica norte para el año 2017Aristizábal Uchima, Harold; Macías Bohórquez, Ricardo; Santana Contreras, Juan CamiloThe time series method has been often used to do forecast in different topics, especially economic topics. One of the most important uses is to support the planning area in all kind of companies and thus achieve anticipation. In this research different models were generated to contrast the inside forecast to company and the forecast generated with statistics methodology. The data have information about agrochemical sales in Latino American countries by a multinational company, since January of 2012 until December of 2016. In the analysis section it was found that all time series had seasonality and for that reason all the models were SARIMA, in some cases outliers were included to these models, and it was generated a forecast to 2017. Finally, it was concluded that statistical forecast models have more precision. but companies cannot rule out empirical knowledge in generating them.
- ÍtemDeterminantes de la producción agrícola en Colombia: Una aproximación aplicando modelos de datos panel espacialesVargas Suárez, Jeymy Carolina; López Ospina, Diego Felipe; Santana Contreras, Juan Camilo
- ÍtemDeterminantes de la reducción de la pobreza en 6 países de américa latina entre los años 2001 -2014: una estimación en datos panelDíaz Buelvas, Eduardo Luis; Santana Contreras, Juan CamiloThe following document seeks to identify the impact of some macroeconomic variables on the poverty rate in 6 countries of Latin America (Colombia, Peru, Ecuador, Brazil, Paraguay and Bolivia) in the period between 2001 and 2014; through an economic model in the data panel, having as a dependent variable the poverty line according to the explanatory variables: unemployment rate, GDP growth, inflation, Gini coefficient, public spending on education and growth rate of the population.
- ÍtemModelo estadístico para establecer la relación entre el indicador de cartera vencida total, la pérdida neta de crédito y el costo de cartera de un crédito de consumo libre inversión otorgado por una entidad bancaria de Bogotá, ColombiaMedina Monroy, Oscar Mauricio; Cárdenas Pinto, Henry; Santana Contreras, Juan CamiloThis paper presents a statistical model to establish the relationship between the total overdue portfolio indicator, the net loss of credit and the cost of credit, for a credit of consumption of free investment granted by a bank of Bogot a (Colombia). The data contain information for a period from June 2011 to January 2015. The di erent statistical tests that were applied to the proposed model suggest that it is consistent. The augmented Dickey-Fuller unit root test allows us to establish that the series net loss of credit, credit cost and total overdue portfolio indicator are not stationary and have the same order of integration. The Engle-granger cointegration test provided su cient statistical evidence on the long-run and short-run equilibrium relationship between net credit loss and the total overdue portfolio indicator and between the cost of credit and the net loss of credit. Using historical data from the total overdue portfolio indicator in the proposed statistical model together with the accounting information contained in the pro t and loss statement, a Monte Carlo simulation was carried out, which allowed the forecast of the cost of credit and the rate of return on assets. With the results obtained it is possible to indicate to the bank what the probability of obtaining minimum pro ts for the product of consumer credit of free investment is.
- ÍtemModelo scoring para recuperar cartera de microcrédito(Fundación Universitaria Los Libertadores. Sede Bogotá., ) Rodriguez Avellaneda, Angelica Lucia; Santana Contreras, Juan CamiloIn the present work the construction of a Scoring model is exposed to determine the probability of payment of the client when it is in default, focused on the sector micro credit, looking for a tool with which the portfolio of in an efficient way, reducing costs and increasing the level of recovery. Finally If this objective is met, there is a 6-month base (April-September 2018), with which 4 different models are generated, which are: logistic regression, regression probit, decision trees and neural networks. The percentage of error in each one of them, resulting in the lower percentage generated by decision trees with 11.3%, said model reflects an adequate adjustment, showing a Successful predictability level. As for the remaining models, none have a percentage greater than 12.5%. Therefore, through logistic regression, it is determined that the guarantee, gender and default at the time of evaluation are the variables with the highest impact on non-payment of customers.
- ÍtemPronóstico de la reserva técnica para una empresa del sector salud EPS-SEM bajo metodología BOX-JENKINS(Fundación Universitaria Los Libertadores. Sede Bogotá., ) Moreno Arévalo, Manuel Alejandro; Melo Contreras, Eddy Aldemar; Santana Contreras, Juan Camilo
- ÍtemPropuesta de un índice SIPSA para el pronóstico de la inflación de alimentos. Evidencia empírica(Fundación Universitaria Los Libertadores. Sede Bogotá., ) Lozano Forero, Sébastien; Santana Contreras, Juan CamiloInflation is a delicate macroeconomic agent to deal with since it directly affects citizens' lives. In the case of the index of variation in food prices, such an effect is much more sensitive, both for companies in the food sector and for the population in general. Thus, it is of interest to be able, with some degree of certainty, to obtain relevant information on the index of variation in food prices. In turn, the SIPSA (DANE unit) is the government agency in charge of dictating the majority supply centers in the country; the prices with which the products must be traded. In the present work, the problem of predicting the in fl ation of food from a SIPSA index proposed by the author is addressed. Two methodologies are addressed for this purpose. Namely: integrated self-regressive models of seasonal moving averages with exogenous variables (SARIMAX), self-regressive structural vector models (SVAR) and error correction vector models (VECM).
- ÍtemPropuesta de un índice SIPSA y su relación con la inflación de alimentos para el caso colombiano: evidencia empírica(2019) Lozano Forero, Sébastien; Santana Contreras, Juan CamiloLa inflación de alimentos es un agente macroeconómico de notable impacto en el costo de vida de losciudadanos y en el acercamiento a la toma decisiones de las entidades que inciden en política monetaria.Por tanto, resulta de interés tener la capacidad de, con razonable grado de certeza, obtener informaciónrelevante de la variación en los precios de los alimentos. Por otro lado, el SIPSA (dependencia del DANE) esel ente gubernamental encargado de recopilar los precios con los que se comercializan los productos en lascentrales de abasto del país. En el presente trabajo, se aborda el problema de aportar a la predicción de lainflación de alimentos a partir de un índice SIPSA propuesto por los autores. Se abordan tres metodologíaspara tal fin. A saber: modelos auto regresivos integrados de medias móviles estacionales con variablesexógenas (SARIMAX), modelos vectoriales estructurales auto regresivos (SVAR) y modelos vectorialesde corrección de errores (VECM). Nuestros resultados indican que las variaciones del SIPSA explicanalrededor del 40 % de la variabilidad de la inflación de alimentos, además de la relación en el corto ylargo plazo. La creación y el uso de tales indicadores, pueden convertirse en herramientas ideales parala planeación y la gestión de estrategias de inversión de portafolio con miras a la tecnificación de laspropuestas de inversión en el mercado de capitales.
- ÍtemPropuesta de un modelo estadístico para el análisis y pronóstico de la tasa de crecimiento de las exportaciones de petróleo y sus derivados para Colombia para el período 2001-2018(Fundación Universitaria Los Libertadores. Sede Bogotá., ) Jaimes Monroy, Luis Felipe; Santana Contreras, Juan Camilo
- ÍtemPropuesta para el cálculo del valor de la póliza SOAT de motocicletas según la región del país(Fundación Universitaria Los Libertadores. Sede Bogotá., ) Briceño Rojas, Edwin Alejandro; Morales Marín, Camilo Eduardo; Santana Contreras, Juan CamiloThis research work seeks to understand the problem that has been presented with the high accident rate of motorcycles throughout the national territory of the country, the increase in the number of motorcycles in recent years and how over time this accident directly affects to the insurance companies that sell the compulsory insurance (SOAT), making the value that each motorcycle owner pays to acquire a SOAT policy, is insufficient to cover the accidents in which this type of vehicle is involved. Therefore, it is necessary to carry out a geographical study in order to determine for each region of the country what would be the estimated cost of the SOAT motorcycle policy (pure premium calculation by region), taking into account the behavior of traffic accidents (value of claims) and the number of policies issued in each region. Given the above, the projection for the year 2018 of the values of the claims and the number of policies expected per region for the motorcycle rate that includes the range of 100 to 200 cubic centimeters, having as input the history of both variables by region of 6 years (2012-2017). This estimate of value by region allowed to determine the excess or defect that exists in the value of the cost of the SOAT policy for this type of motorcycle currently in force in the country, since each region has a different behavior in its exposure to risk which It makes the value differentiating by that variable.