Time series forcasting has been widely used in real life, in which past observations of the same variable are collected and analyzed to develop a model describing the underlying relationship, the model is then used to extrapolate the time series into the future. This modeling approach is particularly useful when little knowledge is available on the underlying data generating process or when there is no satisfactory explanatory model that relates the prediction variable to other explanatory variables.There are several different approaches to time series modeling. Traditional statistical models including moving average, exponential smoothing, and ARIMA are linear in that predictions of the future values are constrained to be linear functions of past observations. In this project, we proposed a novel forcasting model which combined BP(Back Propagation) Networks, SVM(Support Vector Machine) and ARIMA(Auto-Regressive Integrated Moving Average). From three differnt perspectives we captured the subtle relationships by using BP Network, SVM and ARIMA, respectively. Also, a Date profile system was constructed. The model was experimented on the cash demanding record of the ATM in Jiangsu Bank and it achieved promising performance.
================ This project is mainly programmed by Luo Zhengping, also named as Jimmy Lo. and my teammate Liu Tao, we got a lot of beautiful points and help from our honorable tutor Prof.Yang, and personally I want to thank my two best friends Liu Yanjun and Deng Qi for their kindness and precious friendship. All the rights are reserved. If you have any confusion, please contact me from here.
=============