Analysis Of Factors Affecting Local Fiscal Revenue In Hebei Province And Short-term Forecast

Local fiscal revenue is a comprehensive reflection of the state of regional national economic development.Relevant departments must formulate effective fiscal policies and strengthen the supervision and management of local fiscal revenue.Hebei Province is strategically located around Beijing,Tianjin,and the Bohai Sea.Economic development has promoted the growth of local fiscal revenue.Nowadays,with the development of an important strategy for the coordinated development of Beijing,Tianjin and Hebei,Hebei Province has ushered in favorable development opportunities.Precise fiscal revenue forecasting has very important application value in improving the level of decision management and promoting the coordinated and rapid development of the national economy.Establishing a scientific and reasonable forecasting mechanism and proposing more innovative and effective forecasting methods are of great significance for the construction of Hebei economy and the promotion of social development.Based on the analysis of the characteristics and shortcomings of the existing forecasting methods,this paper takes the local fiscal revenue and other relevant economic indicators of Hebei Province from 1994 to 2017 as samples.First,use the Adaptive-Lasso model and the random forest model to choose the factors of local fiscal revenue.Secondly,XGboost regression prediction modeling was performed for each of the variables selected.At the same time,it was compared with the model established without any variable screening.The average absolute error and R square were used as the measurement criteria.The results found that after random variables were selected by the random forest method,the average absolute error of the model fitted by XGboost modeling is smaller than the other two models,and the prediction effect is good,while the R square is larger than the other two models,and the fitting effect is good.Finally,nine index factors affecting local fiscal revenue in Hebei Province were selected based on the random forest method.The gray prediction model was used to predict the values of nine single-factor indicators from 2018 to 2020,and the prediction accuracy was evaluated.At the same time,the BP neural network was established in combination.The network model predicts the local fiscal revenue of Hebei Province from 2018 to 2020,and explores the impact of taxation,industrial structure,total social fixed asset investment,labor compensation,and local fiscal expenditure on local fiscal revenue in Hebei Province.In the following,relevant suggestions for sustainable high-quality development of local fiscal revenue in Hebei Province are put forward.The method of this paper is to make a theoretical in-depth analysis of the relevant forecasting methods,to compare and select the index factors that affect the local fiscal revenue of Hebei Province,and then use the combination model to make short-term forecasts with high accuracy and broad application prospects.This has certain reference significance for Hebei Province to formulate related fiscal policies and achieve long-term economic and social development.