基尼系数的预测建模方法及在北京市经济发展分析中的应用

洛伦兹曲线通常被用于描述社会收入分布的不平均程度。根据洛伦兹曲线可以计算基尼系数。目前．洛伦兹曲线和基尼系数在经济领域中的应用已十分广泛。它们不仅仅被用于收入分配问题的研究，而且已经成为一种有效的均衡分析的统计工具。利用Aitchson提出的对数比率变换，在成分数据预测建模方法的基础上．提出洛伦兹曲线的预测模型和基尼系数的预测方法。作为实证分析的案例研究，利用北京市城八区的人口和GDP数据．对城八区经济发展的不平衡趋势进行推测，得到合理的分析结论．

Predictive Modeling Method of Gini Coefficient and Its Application in Beijing Economic Development Analysis

Lorenz curve is commonly used to describe the inequality degree of social distribution of income. According to Lorenz curve, Gini coefficient is defined as the measurement of inequality of income. Nowadays, Lorenz curve and Gini coefficient have been widely used in the field of economics, which are not only applied on the study of income distribution problem, but also becoming a kind of effective statistical measurement for general equality analysis. Based on the logratio transformation proposed by Aitchison and the predictive modeling approach of compositional data, the paper proposes a predictive modeling method for Lorenz curve and Gini coefficient. As a case study, the economic development of eight urban districts in Beijing has been investigated in the paper. By using the index of population and GDP, the paper forecasts eight urban districts' imbalance trends in economic development, and obtains rational conclusions.