چکیده:
he application of the artificial neural networks in economics and business goes back to 1950s, while the main part of the applications has been developed in more recent years. Reviewing this research indicates that the development and applications of neural network are not limited to a specific application area as it spans a wide variety of fields from prediction to classification, as most of the applications in economics primarily focus on the predictive power of the neural networks. Many researches using statistical and Neural Networks (NNs) models in economics but few involved support vector machines in their studies. In this paper for the first time we compare the approximate economic behavior ability of artificial neural networks (ANN) and support vector machines using a set of data on some Middle East countries.
خلاصه ماشینی:
"A Comparative Approximate Economic Behavior Analysis of Support Vector Machines and Neural Networks Models Amin Gharipour Morteza Sameti 1 Ali Yousefian 2 Abstract T Keywords: Economic Dep.
Through automatic searches for the best indicators for predicting GDP one and four steps ahead, they compared the out-of-sample forecasting performance of adaptive models using different data vintages.
In another area addressing GDP prediction, Tkacz (2001) used neural networks to find out more exact leading indicator models of Canadian output growth by analyzing the forecast performance of multivariate neural networks and finding that there are gains in the short-run forecast precision of the neural networks in comparison to the best linear model due to the neural networks’ ability to capture non-linear relationships in the data.
As there is no structured way to choose the optimal parameters of SVMs, the values of the kernel parameter C and that produce the best result on the validation set are used for the standard SVMs. In terms of using artificial neural network we implement Multilayer Perceptron Network with three layer, 21 neurons in input layer 28 neurons in hidden layer by tan-sigmoid transfer function for predict Middle East GDP.
In this paper for the first time we compared the approximate economic behavior ability of artificial neural networks (ANN) and support vector machines (SVM) model using a set of data on some Middle East countries.
The results showed that neural network out perform support vector machines in both terms of generalization from training data set and accuracy of approximation."