Font Size: a A A

Research On The Influence Factors Of Carbon Intensity Based On Skew-normal Mixed Model

Posted on:2018-02-25Degree:MasterType:Thesis
Country:ChinaCandidate:L J XuFull Text:PDF
GTID:2359330515462894Subject:Statistics
Abstract/Summary:PDF Full Text Request
With the rapid development of China's economy,as well as the process of industrialization and urbanization,the consumption of energy has been increased dramatically.The heavy energy consumption has resulted in large amounts of carbon dioxide emissions,which reflected the serious environmental problem.As China has ranked first in the global carbon emissions,we need not only to undertake international carbon-reduction targets,but also to maintain the rapid economic development.Therefore,many scholars have focused on how to improve energy efficiency,reduce carbon intensity promote low-carbon and sustainable development in our country.However,with the test of significance,the longitudinal data of carbon intensity in different regions of our country doesn't follow the normal distribution.If the traditional econometric methods are still modeling in the research,the results of the statistical analysis will not be robust.So it is particularly important to build the skew-normal distribution model to study the actors of carbon intensity and provide suggestions for reducing China's carbon emission based on the distribution characteristic of actual carbon intensity data.Firstly,the maximum likelihood estimation is presented for the skew-normal mixed effects model based on the EM algorithm.And then,based on the parameter bootstrap,an exact test for the parameter of one-way classification model with skew-normal random effects has constructed.On this basis,we select the data of carbon intensity vary from 2000 to 2014 provincially and testify their skew-normal distribution.And then,a skew-normal mixed effects model is constructed to study the main factors of carbon intensity provincially.In order to illustrate the excellent capability of skew-normal mixed model,we finally compared the above model parameters with the normal mixed model.The result shows that there are several factors influence the carbon intensity in our country,such as per capita GDP,energy consumption intensity,the proportion of the second industry and the ratio of dependence on foreign trade.Specifically,it benefits the reduction of carbon intensity as we increase R&D expenditure to develop technology,reduce the proportion of the second industry and promote the opening policy.It is the most direct and effective way to reduce the carbon intensity by reducing energy intensity and improving energy efficiency.At the same time,China is still in the extensive economic growth model,and we will accelerate efforts to transforms to intensive growth.
Keywords/Search Tags:Skew-normal Mixed Effects Model, Carbon Intensity, EM Algorithm, Parameter Bootstrap, Monte Carlo
PDF Full Text Request
Related items