Predictability and forecasting automotive price based on a hybrid train algorithm of MLP neural network

Abstract

In this paper, using the Hurst exponent value H, we first show that the automotive price in Iran Khodro Company (IRAN) is predictable and therefore a good forecasting can be done using neural networks. We then introduce a new global and fast hybrid multilayer perceptron neural network (MLP-NN) in order to forecast the automotive price. In our new framework, we hybridize the genetic algorithm (GA) and least square (LS) method in order to train the connected weights of the network, which leads us to have a global and fast network. To do so, the connected weights between input and hidden layers are trained by GA and the connected weights between the hidden and output layers are trained by LS method. We finally apply our new MLP-NN to forecast the automotive price in Iran Khodro Company, which is the biggest automotive manufacturing in IRAN. The results are well promising compared with the cases when we apply the GA and LS individually. We also compare the results with the case when we employ the gradient-based optimization techniques such as Levenberg-Marquardt method as well as some heuristic algorithms such as extended tabu search algorithm instead of LS method and hybridization of MLP-LM with GA.

Novel MLP Neural Network With Hybrid Tabu Search Algorithm

“…Motivated by this work, in this paper, our attempt is to introduce another hybridization technique for forecasting the automotive price. The proposed hybrid scheme enjoys a global optima and good performance compared with that presented in [20]. To be exact, our scheme leads us to have a significant reduction in computational time as well as the high accuracy of the obtained solution in comparison with that of [20].…”

“…For this purpose, several attempts have been made in the literature for developing such combinations especially for GA, see e.g. [18,20].…”

“…Iran Khodro Company, it is of our great interest in this paper to use the same line of research in hybridizing local and global heuristic and meta-heuristic methods for accomplishing this job. The first work in this subject was done by the authors in [20] in which a hybridization of GA and LS is introduced for training connected weights in the network. This hybridization enjoys some advantages such as getting rid of falling in the local optima and having a fast convergence rate with low computational results.…”

“…In order to show the efficiency and effectiveness of the proposed hybrid scheme, we apply the scheme for forecasting the automotive price in the Iran Khodro Company. As we are interested in comparing the results with the hybrid technique proposed in [20], we restrict ourselves to the price data obtained in the period of March 2007 to March 2009. Moreover, it has to be noted that, due to the policy of the Iran government in removing subsidies in the market since December 2010, in our opinion, the price data are not reliable enough for citation and, therefore, we ignore those here.…”

Ascertaining Predictability Cognizance for the Prediction of Reference Evapotranspiration

“…Many authors worked on improving the accuracy of prediction in various Artificial intelligence techniques. For instance, C.Sivapragasam and Shie-Yui Liong [10] improved prediction accuracy by identifying optimal training data set. Instead of initiating prediction without any strategy, researchers have also proposed ascertaining the cognizance of predictability before prediction [11].…”