3 resultados para series-parallel model
em Universidade Federal do Rio Grande do Norte(UFRN)
Resumo:
Forecast is the basis for making strategic, tactical and operational business decisions. In financial economics, several techniques have been used to predict the behavior of assets over the past decades.Thus, there are several methods to assist in the task of time series forecasting, however, conventional modeling techniques such as statistical models and those based on theoretical mathematical models have produced unsatisfactory predictions, increasing the number of studies in more advanced methods of prediction. Among these, the Artificial Neural Networks (ANN) are a relatively new and promising method for predicting business that shows a technique that has caused much interest in the financial environment and has been used successfully in a wide variety of financial modeling systems applications, in many cases proving its superiority over the statistical models ARIMA-GARCH. In this context, this study aimed to examine whether the ANNs are a more appropriate method for predicting the behavior of Indices in Capital Markets than the traditional methods of time series analysis. For this purpose we developed an quantitative study, from financial economic indices, and developed two models of RNA-type feedfoward supervised learning, whose structures consisted of 20 data in the input layer, 90 neurons in one hidden layer and one given as the output layer (Ibovespa). These models used backpropagation, an input activation function based on the tangent sigmoid and a linear output function. Since the aim of analyzing the adherence of the Method of Artificial Neural Networks to carry out predictions of the Ibovespa, we chose to perform this analysis by comparing results between this and Time Series Predictive Model GARCH, developing a GARCH model (1.1).Once applied both methods (ANN and GARCH) we conducted the results' analysis by comparing the results of the forecast with the historical data and by studying the forecast errors by the MSE, RMSE, MAE, Standard Deviation, the Theil's U and forecasting encompassing tests. It was found that the models developed by means of ANNs had lower MSE, RMSE and MAE than the GARCH (1,1) model and Theil U test indicated that the three models have smaller errors than those of a naïve forecast. Although the ANN based on returns have lower precision indicator values than those of ANN based on prices, the forecast encompassing test rejected the hypothesis that this model is better than that, indicating that the ANN models have a similar level of accuracy . It was concluded that for the data series studied the ANN models show a more appropriate Ibovespa forecasting than the traditional models of time series, represented by the GARCH model
Resumo:
Amorphous silica-alumina and modified by incipient impregnation of iron, nickel, zinc and chromium were synthetized in oxide and metal state and evaluated as catalysts for the chloromethane conversion reaction. With known techniques their textural properties were determined and dynamics techniques in programmed temperature were used to find the acid properties of the materials. A thermodynamic model was used to determine the adsorption and desorption capacity of chloromethane. Two types of reactions were studied. Firstly the chloromethane was catalytically converted to hydrocarbons (T = 300 450 oC e m = 300 mg) in a fixed bed reactor with controlled pressure and flow. Secondly the deactivation of the unmodified support was studied (at 300 °C and m=250 g) in a micro-adsorver provided of gravimetric monitoring. The metal content (2,5%) and the chloromethane percent of the reagent mixture (10% chloromethane in nitrogen) were fixed for all the tests. From the results the chloromethane conversion and selectivity of the gaseous products (H2, CH4, C3 and C4) were determined as well as the energy of desorption (75,2 KJ/mol for Ni/Al2O3-SiO2 to 684 KJ/mol for the Zn/Al2O3-SiO2 catalyst) considering the desorption rate as a temperature function. The presence of a metal on the support showed to have an important significance in the chloromethane condensation. The oxide class catalyst presented a better performance toward the production of hydrocarbons. Especial mention to the ZnO/Al2O3-SiO2 that, in a gas phase basis, produced C3 83 % max. and C4 63% max., respectively, in the temperature of 450 oC and 20 hours on stream. Hydrogen was produced exclusively in the FeO/Al2O3-SiO2 catalysts (15 % max., T = 550 oC and 5,6 h on stream) and Ni/SiO2-Al2O3 (75 % max., T = 400 oC and 21,6 h on stream). All the catalysts produced methane (10 à 92 %), except for Ni/Al2O3-SiO2 and CrO/Al2O3-SiO2. In the deactivation study two models were proposed: The parallel model, where the product production competes with coke formation; and the sequential model, where the coke formation competes with the product desorption dessorption step. With the mass balance equations and the mechanism proposed six parameters were determined. Two kinetic parameters: the hydrocarbon formation constant, 8,46 10-4 min-1, the coke formation, 1,46 10-1 min-1; three thermodynamic constants (the global, 0,003, the chloromethane adsorption 0,417 bar-1, the hydrocarbon adsorption 2,266 bar-1), and the activity exponent of the coke formation (1,516). The model was reasonable well fitted and presented a satisfactory behavior in relation with the proposed mechanism
Resumo:
Forecast is the basis for making strategic, tactical and operational business decisions. In financial economics, several techniques have been used to predict the behavior of assets over the past decades.Thus, there are several methods to assist in the task of time series forecasting, however, conventional modeling techniques such as statistical models and those based on theoretical mathematical models have produced unsatisfactory predictions, increasing the number of studies in more advanced methods of prediction. Among these, the Artificial Neural Networks (ANN) are a relatively new and promising method for predicting business that shows a technique that has caused much interest in the financial environment and has been used successfully in a wide variety of financial modeling systems applications, in many cases proving its superiority over the statistical models ARIMA-GARCH. In this context, this study aimed to examine whether the ANNs are a more appropriate method for predicting the behavior of Indices in Capital Markets than the traditional methods of time series analysis. For this purpose we developed an quantitative study, from financial economic indices, and developed two models of RNA-type feedfoward supervised learning, whose structures consisted of 20 data in the input layer, 90 neurons in one hidden layer and one given as the output layer (Ibovespa). These models used backpropagation, an input activation function based on the tangent sigmoid and a linear output function. Since the aim of analyzing the adherence of the Method of Artificial Neural Networks to carry out predictions of the Ibovespa, we chose to perform this analysis by comparing results between this and Time Series Predictive Model GARCH, developing a GARCH model (1.1).Once applied both methods (ANN and GARCH) we conducted the results' analysis by comparing the results of the forecast with the historical data and by studying the forecast errors by the MSE, RMSE, MAE, Standard Deviation, the Theil's U and forecasting encompassing tests. It was found that the models developed by means of ANNs had lower MSE, RMSE and MAE than the GARCH (1,1) model and Theil U test indicated that the three models have smaller errors than those of a naïve forecast. Although the ANN based on returns have lower precision indicator values than those of ANN based on prices, the forecast encompassing test rejected the hypothesis that this model is better than that, indicating that the ANN models have a similar level of accuracy . It was concluded that for the data series studied the ANN models show a more appropriate Ibovespa forecasting than the traditional models of time series, represented by the GARCH model
