Font Size: a A A

Application Of The Combination Model In Forecasting The Economic Of China

Posted on:2012-02-26Degree:MasterType:Thesis
Country:ChinaCandidate:X P YouFull Text:PDF
GTID:2219330335476267Subject:Probability theory and mathematical statistics
Abstract/Summary:PDF Full Text Request
Time series forecasting is an important area of forecasting in which past observations of the same variable are collected and analyzed to develop a model describing the underlying relationship.We have a lot of time series forecasting methods,such as neural network and the traditional time series analysis methods,which get advantages in dealing with stationary time series ,to some extent,does not achieve the hope for results.Autoregressive integrated moving average (ARIMA) is one of the popular linear models in time series forecasting in recent years.Statistical Learning Theory(SLT) focuses on the learning theory of small samples.The core of the theory is to control the generalization of learning machine by controlling the complexity of models.Support Vector Machine(SVM)is a general learning algorithm developed from SLT.Support Vector Regression (SVR)is the expansion of SVM to regression problems.SVM is a method of machine learning based on structural risk minimization principle of the statistical learning theory,using kernel function to solve classification and regression problem.In recent years,support vector regression(SVR),a novel neural technique, has been successfully used for financial forecasting.This paper mainly discusses the performance of using combination model to forecast China's GDP.Using hybrid model or combining several models has become a common practice to improve the forecasting accuracy .Combination of forecasts from more than one model often leads to improved forecasting performance.Most of the individual models evaluated showed poor ability to detect directional change,making use of the favorable specific property that combined models may put forward to the forecasting system,the paper sets up the model which combined exponential smoothing model,ARIMA model and SVR model can improve the accuracy of fix and forecast by proper weighs.Because ARIMA and AVR are often compared with mixed conclusions in terms of the superiority in forecasting performance.Then a hybrid methodology that combines both ARIMA and SVR models is proposed to take advantage of the unique strength of ARIMA and SVR models in linear and nonlinear modeling.Comparing with those,a hybrid methodology that combines both exponential smoothing model and SVR model.Experimental results with real data sets indicate that the combined model can be an effective way to improve forecasting accuracy achieved by either of the models used separately.
Keywords/Search Tags:ARIMA, SVR, exponential smoothing, Combination model
PDF Full Text Request
Related items