Font Size: a A A

China's Money Supply Factor Decomposition Model And The Arima Model Forecasts

Posted on:2010-11-20Degree:MasterType:Thesis
Country:ChinaCandidate:Y X SunFull Text:PDF
GTID:2199360305993511Subject:Probability theory and mathematical statistics
Abstract/Summary:PDF Full Text Request
In 2008, a global financial turmoil triggered by the United States sub-prime crisis, which is ferocious, wide and deep impact, has brought heavy losses to every country. An important warning from this is that a country's money supply has to fit its real economy. Therefore, it is really essential to analyze the main factors to China's money supply, to predict accurately the trend of the money supply, and to achieve the expected macroeconomic objectives by reasonable control of the money supply.This paper consists of two parts. The first part is the money supply's factor decomposition, and the second part is the money supply's time series prediction.In the analysis of the factors to money supply, the thesis uses four economic indexes including GDP, price index, funds outstanding for foreign exchange and balance of government bonds. Firstly, we use Granger causality test to show that all the four variables have influence on the alteration of M2. Then we take a co-integration test which removes the price index, and indicate that the money supply has a long-term stable relationship with the GDP, balance of government bonds and funds outstanding for foreign exchange. And also, from the error correction model we know that all the three left factors have positive effects, among which the most important factor is GDP, followed by the balance of government bonds and finally funds outstanding for foreign exchange. The coefficients are 0.76,0.25, and 0.048 respectively. Furthermore, we set a VAR model, which forecast that the M2 of 2009 would be 59.3397 trillion Yuan. And we do an impulse response function analysis and variance decomposition based on the VAR model, which show that GDP has the greatest impact.As for the time-series forecasting, we use the monthly data from January,2000 to December,2008 to fit the prediction model through logarithm, first-order difference and seasonal difference. Its average relative error is 1.84%, and its predicted money supply of December, 2009 money supply is 59.31058 trillion Yuan, which is close to the 59.3397 trillion Yuan of the VAR model. Finally, I gave some policy recommendations to reasonable money supply from the perspective of funds outstanding for foreign exchange, debt and money supply.
Keywords/Search Tags:money supply, co-integration test, VAR model, ARIMA model
PDF Full Text Request
Related items