Several Forecasting Models Of Urban Registered Unemployment Rate In Shandong Province

The issue of "employment" is a topic of pro found influence both in China and in the whole world.At the economic symposium,Premier Wen Jiabao once said that "we will intensify efforts to promote employment,so that the broad masses of people have jobs and enough food to eat." Employment is a major issue concerning people's livelihood.The country attaches great importance to the promotion of employment.On the one hand,this reflects the country's concern for people's livelihood.On the other hand,it also reflects that the current employment situation in our country is not optimistic.Therefore,it is particularly important to forecast the employment situation.Employment is usually measured by the unemployment rate.At present the index that our country uses to measure unemployment rate is urban registered unemployment rate.This paper forecasts the urban registered unemployment rate successively from the perspective of time series and regression analysis.The main work is as follows:Firstly,the traditional ARIMA model is established,and the optimal parameter structure of the model is determined according to AIC criterion.The linear function of the current urban registered unemployment rate and its lag period in Shandong Province is established,and the simulation calculation and prediction are carried out by R software.Second,select three industrial output value,consumer price index,commodity retail price index and other indicators related to the urban registered unemployment rate of Shandong Province to make regression prediction.Firstly,principal component analysis was carried out to remove the multicollinearity of the impact factors,and then the impact factors obtained from the principal component analysis were used for support vector regression to obtain the predicted results.Thirdly,on the basis of the second model,the idea of time series is introduced,and a prediction method with a wider prediction range is obtained.Finally,the conclusion is drawn.The first method is simple and easy to operate,and has a far prediction range,but the prediction accuracy is low.The second method is more accurate,but can only predict the unemployment rate for one year.However,the third method has a better comprehensive performance in the prediction accuracy and prediction range.