Forecast And Analysis Of National Fiscal Revenue Based On ARIMAX Model

State revenue is the amount of funds raised by a government to carry out government functions,public policies,and provide public goods and services.Fiscal revenue is not only the basic guarantee and precondition of government financial expenditure,but also an important index reflecting the wealth level of a country and a government.With China’s socialism entering a new era,our country’s financial work is facing a series of rare development opportunities and grand challenges.Fiscal work plays an increasingly prominent role in strengthening national macro-control,but on the other hand,it is a short board that needs to be improved urgently.Under such circumstances,the establishment of an effective financial revenue forecasting system is becoming more and more important.As an important branch of statistics,time series analysis is widely used in the field of economic prediction.It could find an optimal fitting model by analyzing and studying a set of time series data,and could also be used to predict its future trend changes.It has a strong practical significance in our daily life.At present,research on time series analysis is rare in fiscal revenue,and it basically focuses on the field of unitary time series.In this paper,we use the national fiscal revenue data from 1978 to 2013 as the original data of the model,and use the GDP data as the covariate to establish the multiple dynamic regression model—ARIMAX model.This model is used to predict the change trend of national fiscal revenue in the next five years.The results are compared with the results of a classical one-dimensional time series model(ARIMA),the real national fiscal revenue data from 2014 to 2017.The experiments show that ARIMAX model is not only feasible,but also more accurate and effective than the classical one-dimensional time series model.The ARIMAX model is a superior time series model and can be applied to the analysis and prediction of national financial revenue.