The Research Application Of Education Investment And Economic Growth Base On Time Series Analysis

In the actual life,people need to observate some phenomenon for explorating the rules ofcertain things or systems,that get to a certain sequence of data. And the time series analysisbased on data analysis predicts or controls the development of things or systems in thefuture.In practice,most sequences are not stable,linear.Time series analysis uses the form oflogarithmic difference to smooth the data, reaching the expected effect through the modelfitting and prediction,which represents the data that reveals the regularity and developmenttrend of the nature of things.This paper mainly analyses the relationship of education investment,GDP and thenational financial expenditure. Education as an intangible asset can improve people’s culturallevel and the labor productivity, promote human capital accumulation and development,encourage the progress of science and technology, thus get forward the whole social economicdevelopment and the adjustment of economic structure and optimization. In turn, the economyis the material basis of the growth on education investment and economic growth willenhance the social investment.Therefore, more and more countries take education investmentas the economic sustainable development of the power.This paper introduces the background of introduction,the present situation ofdomestic,foreign scholars research and the paper’s main structure.The second chapter presentsthe basic concepts of the time series and gives the significance of sequence stability. The thirdchapter introduces the AR,MA,ARMA,ARIMA model and discusses the effect of the model.The fourth chapter basically introduces the multiple cointegration analysis and the errorcorrection model for analyzing data of long-term and short-term effect. The fifth chapter isempirical analysis and separately sets up the fitting model based on the data of our countryeducation expenditure,GDP, national financial expenditure.Through the comparison analysis,we can conclude that the error correction model based on the multiple cointegration analysiscan simulate and predict data well. The sixth chapter is summary and prospect, get theconclusion of this paper and also put forward the improvement,the study of space.In the empirical analysis we describe the origin of data, introduce the experiment of dataprocessing, set a model fitting and forecast the development trend. The last,we compareARIMA model and the error correction model based on multiple cointegrationanalysis.Therefore,we get the conclusion: the multiple error correction model can be better tofitting and prediction.