模拟1947 - 2005年USA个人收入基尼系数的演变
Modeling the evolution of Gini coefficient for personal incomes in the USA between 1947 and 2005
摘要：
对1947 - 2005年美国个人收入基尼系数的演变进行了分析和建模。美国人口普查局（US CB）为这一时期提供了几个不同分辨率的个人收入分布（PID）版本。实际上，由于离散和连续表示之间的差异，这些PID导致不同的基尼系数。如果将所有15岁以上的人都列入人口发展指数，基尼系数从1947年的0.64下降到1990年的0.54。这种影响是由于没有收入的人的比例大幅下降而观察到的。对于不包括无收入者的PID，基尼系数在1960年至2005年期间在0.51左右变化，标准差为0.004，即实际上是恒定的。根据收入定义的任何修订，这一基尼系数实际上与收入指数所包括的人口比例无关。描述个人收入（微观经济层面）及其总价值（宏观经济层面）演变的模型的驱动力是人均名义国内生产总值的变化。该模型准确地预测了基尼系数的演变为PID的人的收入。该模型给出了1947年至2005年期间几乎不变（归一化）的PID和基尼系数。当没有收入的人数减少时，经验基尼曲线会向预测的曲线收敛。当所有的劳动年龄人口都获得了一个适当的收入定义时，经验曲线将逐渐向理论曲线塌陷。因此，模型基尼系数可能更好地描述了美国不平等的真实行为，因为美国人口普查局使用的收入定义显然未能描述真实的收入分配。
英文摘要：
The evolution of Gini coefficient for personal incomes in the USA between 1947 and 2005 is analyzed and modeled. There are several versions of personal income distribution (PID) provided by the US Census Bureau (US CB) for this period with various levels of resolution. Effectively, these PIDs result in different Gini coefficients due to the differences between discrete and continuous representations. When all persons of 15 years of age and over are included in the PIDs, Gini coefficient drops from 0.64 in 1947 to 0.54 in 1990. This effect is observed due to a significant decrease in the portion of people without income. For the PIDs not including persons without income, Gini coefficient is varying around 0.51 between 1960 and 2005 with standard deviation of 0.004, i.e. is in fact constant. This Gini coefficient is practically independent on the portion of population included in the PIDs according to any revision of income definitions. The driving force of the model describing the evolution of individual incomes (microeconomic level) and their aggregate value (macroeconomic level) is the change in nominal GDP per capita. The model accurately predicts the evolution of Gini coefficient for the PIDs for people with income. The model gives practically unchanged (normalized) PIDs and Gini coefficient between 1947 and 2005. The empirical Gini curves converge to the predicted one when the number of people without income decreases. Asymptotically, the empirical curves should collapse to the theoretical one when all the working age population obtains an appropriate definition of income. Therefore the model Gini coefficient potentially better describes true behavior of inequality in the USA because the definitions of income used by the US Census Bureau apparently fail to describe true income distribution.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
