Title :
Financial time series forecasting based on v-SVRNN
Author_Institution :
Coll. of Quality & Safety Eng., China Jiliang Univ., Hangzhou, China
Abstract :
In this paper briefly introduces the basic theory of Support Vector Regress (SVR), and applies V-SVR combined with neural network (V -SVRNN) to create a model, which also can be used for forecasting the financial time series. Different input variables, multi-step prediction and one-step prediction are studied in this paper. The results of simulation show that the new model is the least in the mean squared error, which demonstrates that the V-SVRNN model has a good ability to generalize.
Keywords :
economic forecasting; finance; mean square error methods; neural nets; regression analysis; support vector machines; time series; V-SVR combined with neural network; financial time series forecasting; mean squared error; multistep prediction; one step prediction; support vector regress; Artificial neural networks; Computer languages; Educational institutions; Forecasting; Input variables; Support vector machines; Time series analysis; Data mining; Support Vector Regress; financial time Series; neural network;
Conference_Titel :
Electrical and Control Engineering (ICECE), 2011 International Conference on
Conference_Location :
Yichang
Print_ISBN :
978-1-4244-8162-0
DOI :
10.1109/ICECENG.2011.6057299
