Optimizing Social Economy Prediction based on Integration of Time Series Models

Under the guidance of the reform and opening-up policy, China's economy has developed at a high speed for about 40 years. Fiscal revenue is not only the symbol of a country's economic strength but also an important indicator to evaluate the financial resources of a country's government as it can be driven by the economic growth. However, China's economic growth has begun to slow down in recent years due to unstable international relationships although its the second largest economy in the world. Therefore, it's imperative for us to explore future development of China's fiscal revenue to take corresponding reactions in advance. The objective of this paper is applying SARIMA model and LSTM model to forecast China's fiscal revenue. The impact of inflation is taken into consideration when studying economic data, the consumer price index (CPI) is used to eliminate the impact of it. This is the first time that the impact of inflation is taken into consideration to forecast China's fiscal revenue and successfully achieve better performance.