“中等收入陷阱”的贝叶斯网络推断研究

Study on Bayesian NetworkInference of“Middle-Income Trap”

关于中国是否进入中等收入陷阱存在很大的争论,根据中等收入陷阱产生的根源、表征等构造出一个贝叶斯网络模型,通过世界各地的对应数据,按照极大似然法对模型的各个参数进行估计,从而得到中等收入陷阱的基本规律,并根据这一先验规律对中国的情况进行贝叶斯推理。研究结果发现,中国不会发生中等收入陷阱,其发生概率主要受到低端制造业转型的影响,同时产业在发展过程中要高度重视技术创新、发展公平性与体制机制改革。随着时间的推移,七大经济区域走出中等收入阶段向高收入迈进的顺序是东—中—西依次迈进,即华东地区向华中、华北、华南、东北地区再向西南、西北地区,区域之间存在相互传染、相互扩散。

There is a great controversy about whether China has entered the middle-income trap.This paper constructs a Bayesian network model based on the origin and symptom of the middle-income trap.Through the corresponding data from all over the world,the parameters of the model are estimated according to the maximum likelihood method,and the basic law of the middle-income trap is obtained.According to this prior law,the situation of China is also discussed Bayesian reasoning.The results show that:(1)There will be no middle-income trap in China,with a maximum probability of 48%,which is mainly affected by the transformation of low-end manufacturing industries,while the industry should attach great importance to technological innovation,development equity and institutional mechanism reform in the process of development;(2)As time goes on,the order of the seven major economic regions moving out of the middle-income to high-income is as follows:East-middle-west is advancing in turn,that is to say,East China is moving towards central,north,South and northeast China,and then to southwest and northwest China,there is mutual contagion and diffusion between regions.