Forecasting Of Per Capita GDP Of China That Based On Combination Forecasting Method

In recent years, GDP has been the social focus of attention because it is one of the important indicators to measure a country’s economic situation, which plays an important role in reflecting the State of a country’s economic development, national income and consumption capacity. The per capita GDP not only take into account the size of a country’s economic output, but also combines the factors of population, so it is widely used in the international evaluation of a country’s or region’s affluence and the level of people’s living standards. For economy stage of rapid development of China’s, timely and accurate forecasts on the per capita GDP is the national important prerequisite for effective regulatory policy.According to the characteristics of the economic analysis and based on summarizing previous research on combined forecasting method and per capita GDP forecasting method, this paper introduced combined forecast method into China’s per capita GDP forecast application in order to increase the accuracy of per capita GDP forecast. Specific completion of work and research include:(1) Researched on per capita GDP for short-term forecasting analyses based on ARIMA model, and detailed description of the modeling steps and used for analysis’and prediction’s features of ARIMA model.(2) Researched on per capita GDP prediction analysis based on exponential smoothing model. Described the characteristics of exponential smoothing models for analysis and prediction.(3)Through the determining method of combined weight,we established combinaed forecasting model based on residual reciprocal method, combinaed forecasting model based on the inverse of the variance method and the combinaed forecasting model based on least squares method.Compared the various forecasting models and prediction by comparing effect evaluation indicators, found three combined prediction model’s accuracy significantly better than each individual forecast model. This illustrates that the combined forecasting model for forecasting China’s per capita GDP is effective. Using three kind of combined forecasting model to forecast per capita GDP of2008to2012, the results are very close to the actual value.The forecast result of countdown variance portfolio model is better than other two combined model. Experimental results show that the proposed combined forecasting method with good accuracy and feasibility.