Tendency analysis and forecast based economic time series data: An analysis of Guangzhou’s total volume of retail sales of social consumer goods

Using regression and statistical hypothesis testing, we investigate trend growth of Guangzhou’s total volume of retail sales of social consumer goods, the results show that Guangzhou’s total retail sales of social consumer goods growth have an accelerating trend and its seasonal fluctuations tend to slow. We also consider the forecasting analysis on Guangzhou’s total retail sales of social consumer goods. The method used for forecasting are the season ARIMA model to consider the impact of the Spring Festival, and Support Vector Machines (SVMs) method. Both methods all give good forecasting accuracy. Relatively speaking, SVMs need less the statistical assumptions on the data that it may be better to avoid the problem of model misspecification. In addition, SVMs can also provide higher accuracy of prediction than the season ARIMA model.