基于人工神经网络的我国宏观经济预测研究-政治经济学专业论文.docxVIP

山东大学硕士学位论文ABSTRACT 山东大学硕士学位论文 ABSTRACT Macroeconomy system represents the comprehensive economic status in a countryor an area．The macroeconomy position need to be predicted when the nation adjusts and controls the economy and the individuals make policy to invest．Prediction is needed to be made at first，which means making the judgment for the trend of the matter through theory and certain methods according to historical records and current conditions．Among the current prediction methods，the most common statistics methods are time—sequencing and return predictions．And macroeconomy is a nonlinear system, whose situation keeps changing．Besides．additional interference factors have direct effects on the operation of macroeconomy systems，greatly influencing the prediction results．it is necessary to solve such problems by using the traditional prediction methods． Because the prediction of Macroeconomy is a multi—variables and non—linearity question，in this paper，artificial Neural networks are applied to predict the macroeconomy． The ANN is an intellective model based on the physiology．It simulates the thinking ability of human’s cerebrums to find the rules by continuous learning，memorying，induceing and fixing of generous cases．The ANN has the ability of collateral operation，non—linearity processing，self-organizing，self-learning and self-adaptation．Since Lapedest and Farber used the ANN to predict firstly．the experts of the inland and overseas continually conjoin it and economics to predict the macroeconomy，strategic economy and stock market and so on． In this paper，we found an artificial neural network model to predict our national macroeconomy position．We use the nth year’s final consumption expenditure，total capital formation and net export of goods and services to predict the next year’s GDP．Therefore，in the end of the nth year，the people could predict the next year’s GDP，so they could estimate the economy position and make the best economy decision．We use VC+