Research On Effective Calculation And Combination Forecast Of Urban Unemployment Rate

“Employment”has always been a major issue of national concern.The blue book of society released by the Chinese academy of sciences in 2009,the “six stability "" stressed at the central economic work conference held in Beijing in December 2018,and the government work report delivered by premier li keqiang in March 2019 all highlighted the focus on employment.Employment is the foundation of people's livelihood,and is closely related to the lives of thousands of families.The employment situation is usually reflected by the unemployment rate,and the unemployment rate is also a reference for the relevant departments to make relevant policies.According to the current official urban registered unemployment rate in China,it cannot accurately reflect the real unemployment situation in China.In order to earnestly implement xi jinping thought on socialism with Chinese characteristics for a new era,fully implement the important decisions and arrangements of the CPC central committee and the state council on employment and employment statistics,and accurately and timely reflect the national employment and unemployment situation,it is necessary to estimate the relatively true unemployment rate.Therefore,it is of great theoretical and practical significance to calculate and forecast the urban registered unemployment rate effectively and scientifically.At present,there are three main methods to calculate the urban registered unemployment rate: adjustment coefficient method,sampling survey method and population factor survey method.In this paper,the first “adjustment coefficient method” is selected to adjust the urban registered unemployment rate.Since the original adjustment coefficient did not consider the influence of population change on the unemployment rate when it was defined,and the adjustment coefficient of the census year was used to replace the adjustment coefficient of adjacent years,the adjustment formula of the urban registered unemployment rate was improved based on the formula of the adjustment coefficient of the urban unemployed population.The improved urban registered unemployment rate adjustment coefficient makes up for the two deficiencies of the original adjustment coefficient.Finally,the selected urban registered unemployment rate from 1990 to 2018 is calculated and the revised urban registered unemployment rate is obtained.Based on the revised urban registration unemployment data,three single prediction models are firstly established,which are the multiple linear regression model based on principal component analysis(pca),the polynomial prediction model and the grey prediction model based on the initial value correction.Secondly,in order to improve the prediction model of the fitting degree of the unemployment rate,on the basis of three kinds of single forecasting model,the traditional gray trend relational degree is extended,the expression of generalized weighted average,the change rule of optimization of average expression form,and then build a new based on the generalized weighted averageunemployment rate of gray trend relational degree of fixed weight coefficient combination forecast model.Finally,the generalized induced ordered weighted average operator(GIOWA operator)is introduced to construct a combined prediction model of variable weight coefficient unemployment rate based on GIOWA operator and grey trend correlation degree.In the grey trend correlation degree variable weight combination prediction model of GIOWA operator,the three special values of GIOWA operator ?(28)1???0 and ?-(28)1are selected to analyze the unemployment rate prediction.Using the lingo software to solve the model,the combination forecast model under different optimization criteria weight coefficient,then calculate combination forecast,and from the redundant forecasting method,curve fitting and prediction precision of the model in the face of combination forecast model and single forecasting model,the fixed weight coefficient of combination forecast model and variable weight coefficient of combination forecasting model between comprehensive evaluation analysis.It is concluded that the combined prediction model of fixed weight coefficient has better prediction effect than the single prediction model,and the combined prediction model of variable weight coefficient has better prediction effect than the single prediction model and the combined prediction model of fixed weight coefficient.