Font Size: a A A

Prediction And Analysis Of The GDP Of Shandong Province By ARIMA Model And ARIMAX Model

Posted on:2017-05-25Degree:MasterType:Thesis
Country:ChinaCandidate:C C ChenFull Text:PDF
GTID:2309330485982238Subject:Applied statistics
Abstract/Summary:PDF Full Text Request
GDP of a country or region is monetary measure of the value of all final goods and services produced in a given period. Hence, GDP can not only be commonly used to show the economic performance of a whole country or region, but provide a reference and basis for macroeconomic policy making. In addition, the GDP is relatively easy to count up because of statistics more accurate and calculation smaller repeated. Since, GDP and economic growth, inflation and unemployment rates of the main macroeconomic indicators have a very close relationship, it is the most fundamental indicators. In short, GDP can show a complete picture of a country or regional economic conditions and help predicate the economic development trend and direction.Therefore,. fore-casting GDP is very necessary under the background of the China’s economic developmentThis paper first describes the background of the topic selection and the present situation of the time series.The theory of time series forecasting is introduced in chapter two. First, it focuses on ARMA model and ARIMA model of one variable time series, however, there are many factors, which have different effects on the predicting in real life, so the next we introduce ARIMAX model of the multivariate time series.In chapter 3 and chapter 4, the empirical analysis has been carried out, we use the GDP data and the output value of the tertiary industry of Shandong Province from 1975 to 2013 as the research object, base on the theory of time series analysis of ARIMA model and ARIMAX model, use SAS software and the model match after data processing, stationarity test, model identification, parameter estimation and model test. From the fitting results, we can see that the predicted value and the true value match well and the true value falls within the 95%confidence interval, and the relative error between the predictive value and the true value is less than 2%, which fully indicate that the model has good fitting effect and high prediction precision. Through the empirical analysis it shows that ARIMA model and ARIMAX model can be applied in practical work for short-term macroeconomic forecast.We use the third industrial output value as the input variables of ARIMAX model, mainly want to analyze the industrial structure of Shandong Province.The innovation of this paper is the introduction of the tertiary industry output value as the input variable of GDP sequence, and the use of multivariate time series data fitting. Compared with the single variable time series, the multivariate time series is more complicated. Finally, through the AIC and SBC criteria, we can see the ARIMAX model is better than the ARIMA model, so it can be inferred that we can add more input variables to predict the value of GDP more accurately.
Keywords/Search Tags:ARMA Model, ARIMA Model, ARIMAX Model, GDP Fore- cast, Industrial Structure Optimization
PDF Full Text Request
Related items