Dissertation > Excellent graduate degree dissertation topics show

Statistical Regularities of City-Size Distribution and Its Formation Mechanism

Author: XueDanZhi
Tutor: GongXiaoQing
School: Zhejiang Technology and Business University
Course: Statistics
Keywords: Power-law Zipf law Pareto distribution Preferences attachment Bounded Rationality Growth mechanism
CLC: C921
Type: Master's thesis
Year: 2010
Downloads: 139
Quote: 0
Read: Download Dissertation

Abstract

The city is a gathering place for a variety of socio-economic factors, forms and conditions of the urban economic survival material space. Currently mentioned that the city size distribution generally think of Zipf's law from another angle, it reflects the power law between objective data. Reflect the city-bit sequence - the Zipf theoretical relationship between the scale has become an important foundation for studies of urban systems, and has been the focus of attention of researchers in the field of. In this paper, the distribution of city size as the object of study to study the formation mechanism of the Zipf law. Established urban population migration space progressive model based main lines, and contains a number of sub-models. The submodel common is that they are used to the the simulated urban population migration model, have the same operating environment, the biggest difference is that the model of the main migration rules. The model is based on the main method largely abandoned the rational choice paradigm that emphasizes participation in the main body of the interaction is limited rationality. This model is a program through the establishment of the model, based on the distribution of urban system in the main space, not to study the formation of a single city, it is the interaction of macroeconomic phenomena emerged through the main micro is a bottom-up model. trying to use a computer simulation of population migration behavior, use this method to study the distribution of the size of the bit sequence of the urban population and its mechanism. The first model is a pure random walk model, city size distribution of approximately obey the same normal distribution. The second model added migration rules the main select the largest city in its field of view as a migration destination, the experiment found that the system soon reaches a steady state and generates the Zipf law. Further verify the following conclusions, \more specific requirements, the need to obey the distribution of the main vision is meant herein is symmetric. The third model of negative growth factors, i.e.-c ??· Nj (t) 2. The experimental results confirmed that gradually decreases as c increases, the size of the largest city. And a growing number of urban growth in scale phenomenon. Scale to achieve the utility function after it is no longer the main added, there will not be the subject moves away, this is not an aggregate actual, also led directly to several of the same size as the largest city in the run results, and therefore continue to optimize the model, to modify the rules of migration, geographic spillovers in the fourth model, found that the goodness of fit is greatly improved, making it possible the emergence of new cities, and the main differences is that the subject's vision does not need to obey symmetric distribution analysis algorithm of the model, city size distribution is Zipf law, and to adjust the main their vision, you can get different Zipf index, which is also found in the data collapse phenomenon, reflecting the Zipf law distribution scale-free type. Intersubjective interaction fifth model, the migration rules closer to human migration rules, there has been found in the urban small-scale deflection. General, at the left end of the distribution of empirical data to meet the often non-power law, the power-law behavior when, that is, when the distribution is beginning to show the form of a power law, it is very important. This paper briefly describes the minimum two estimates BIC criteria and KS statistic, removing less than the minimum value of the scale maps presented Zipf law. Added in the last expansion model growth mechanism, verify the necessity of the growth mechanism Zipf law on the formation time dimension, the experimental results show that the new principal added increases the number of cities in the region and the size of the largest city, but The speed of the increase in the number of cities is less than the speed of the increase in the size of the largest city. In the last part of this paper is the empirical research, Gabaix think that when the population growth rate will not exceed the growth rate of the number of cities, the Zipf exponent be possible to achieve 1. Let's look at the results of the condition does not meet the case this conclusion. Large-scale distribution of the population moving into the province of Zhejiang Province as the research object, the data selection of the 1990 and 2000 census data to study the situation of a speed greater than the increase in the number of cities the size of the largest city to increase the speed The under city size distribution is not to obey Zipf law. Zhejiang Province city size distribution follows Zipf law, and careful analysis and analysis of the characteristics of their city size distribution, and thus to facilitate the government to formulate effective policies and measures to make the city an appropriate scale development, and the environment, the coordinated economic development, sustainable development of road.

Statistical Regularities of City-Size Distribution and Its Formation Mechanism

Abstract

Related Dissertations