Modification And Application For M Function Method Of Industrial Agglomeration

Since the 1990 s of the last century,industrial agglomeration in space has increased the productivity of factors,which in turn has led to the growth of the national economy and has become an important driving force for industrial and economic development.Therefore,related research has also attracted widespread attention from industry and academia.One of the key issues is how to accurately measure the level of industrial agglomeration.From the early GINI coefficient and Herfindal index to the later developed DO index and M function based on geographic information,the accuracy of the measurement is getting higher and higher.However,some methods still have shortcomings.For example,the M function method does not deduct the number of irrelevant companies when measuring the agglomeration trend,resulting in results that are not the actual degree of agglomeration of the industry.In this paper,the M function is modified algorithmically,a new M-Revised function method is proposed,and the advantages of the M-Revised function method are compared and analyzed through simulation experiments.Finally,an example is used to verify the micro data of Suzhou enterprises.The study found that the cumulative function used by the M function makes it more misjudged when measuring the concentration of certain industries.When the industry distribution is characterized by a single point of agglomeration,the M function has the risk of overestimating the level of agglomeration.The modified M-Revised function can better avoid this and more accurately reflect the actual level of agglomeration.When the industry presents multi-point aggregation and random distribution characteristics,M functions(including the DO index)cannot identify the true distribution state.The M-Revised function will report multiple obvious breakpoints,rejecting the assumption that the industry has continuous agglomeration,so as to obtain a more accurate judgment.The use of real corporate microdata also supports conclusions derived from simulated data.In addition,compared with the M-function method,the M-Revised function method can also see the changing range of the degree of industry agglomeration,which is more conducive to improving the accuracy of judgment.The M function and M-Revised function are used to measure the concentration of enterprises in Suzhou,Suzhou Industrial Park,foreign-invested enterprises,and innovative enterprises.The research shows that the results reported by the M-Revised function are consistent with the actual situation and the agglomeration theory.When measuring the agglomeration degree of Suzhou enterprises,as far as the results presented by the M function are concerned,the distribution status of Suzhou enterprises has changed from scattered to neither dispersed nor agglomerated.It cannot be judged that this transition is caused by the cumulative nature of the M function.The measurement error is still the state of the enterprise distribution itself.In addition,empirical results show that when measuring agglomeration,multiple methods,such as the M-function method and the DO index method,should be used for a comprehensive review,which is more conducive to a comprehensive analysis of the characteristics of agglomeration.