Font Size: a A A

Macroeconomics Factors Of Stock Market Fluctuates And Prediction Of Stock Market Price Based On Empirical Model Decomposition

Posted on:2014-02-16Degree:MasterType:Thesis
Country:ChinaCandidate:C L JiFull Text:PDF
GTID:2269330425492893Subject:Quantitative Economics
Abstract/Summary:PDF Full Text Request
For China’s market, after20years of development, it made great achievements. Due to all kinds of contradictions in the process of marketization and the particularity of China’s economic and financial system transition phase, China’s stock market is not yet mature, One of the outstanding performance is that the abnormal fluctuation of stock price phenomena is obvious. A standard of judgment of the securities market is mature or not is the characteristic of volatility, volatility is an significant element of market price behavior. To make the Chinese stock market investors take the right choice keep our country’s economy in long-term, stable, and healthy development, we must take a thorough research on the factors of Chinese stock market volatility.Stock market is an important part of the national economy, which serves it. Stock market volatility and macroeconomic has a close relationship. But due to the stock market has its own laws, economic fluctuations do not agree with stock market volatility.Developing a set of effective theory of the impact of stock price index fluctuation is helpful to guide our country’s macroeconomic policy, so as to promote the development of capital market.It is alse advantageous to guide our country’s macroeconomic policies, so as to promote the development of capital market scientifically.With the development of Chinese economy, the stock market has been gradually improved. More and more people are involved in the stock market and hope to be able to predict the stock market accurately, In practice, If the investors can effectively predict the evolution of the stock index and stock market volatility in other economic variable, then they can take action in advance to avoid risk and to maximize investment profits.This thesis proposes a method about the macroeconomic elements of the stock price volatility based on EMD(empirical mode decomposition method). First, stock prices sequences and macroeconomic variables sequences are decomposed into some components by EMD. Secondly, combining these components, the integration theory, Granger causality test are used to further study the low frequency component. Finally, in full text summary, according to the research conclusion, we bring up the suggestion of some policies.It is indeed challenging to predict stock price because stock price series are non-linear and non-stationary. In order to overcome the limitations of traditional econometric models, artificial neural network, support vector machine (SVM), and genetic algorithm, such as artificial intelligence methods are used to forecast, and the method is better than the traditional models. The paper puts forward a nonlinear combination forecast based on EMD and BNNs.The stock price sequence is decomposed into some components by EMD. Different BNNs models are constructed for every sequences mentioned above to predict the final value respectively. And the final predictive value can be obtained by constructing combined model using the predictive values of every sequence.The empirical results shows that the development of Chinese stock market and macroeconomic is adaptive, the stock price index to a certain extent can reflect the overall level of Chinese macroeconomic development and trends. The industrial added value growth rate, money growth rate has a positive impact; the inflation rate, market interest rates and the exchange rate has a negative impact. The conclusion is basically consistent with the theory. In addition, by comparing to BNNs model, this method in predict the stock price has higher prediction accuracy.
Keywords/Search Tags:EMD, stock market index, macroeconomic variables, predict
PDF Full Text Request
Related items