data field On the contrary, in systems where the information propagates with a finite speed—these systems are intrinsically relativistic—it results κ 6= 0, so that the exponential tails become fat according to Eq. (2a) and the Pareto law emerges

李丽娜

首都经济贸易大学马克思主义学院党总支书记，教授，硕士生导师

 形势与政策教研室» 正文

[PDF]pdf - arXiv

arxiv.org/pdf/1209.4787

• Cached
• Similar

arXiv

Loading...

by F Clementi - ‎2012 - ‎Cited by 8 - ‎Related articles

Dec 7, 2012 - American household wealth surveys—the Survey of Consumer Finances .... wealth are partly caused by the presence of long and heavy tails in the ...... P. Quarati, Kinetic model for q-deformed bosons and fermions, Physics.

 

 

 On the contrary, in systems where the information propagates with a finite speed—these systems are intrinsically relativistic—it results κ 6= 0, so that the exponential tails become fat according to Eq. (2a) and the Pareto law emerges

 

The generalized exponential represents a very useful and powerful tool to formulate a new statistical theory capable to treat systems described by distribution functions exhibiting power-law tails and admitting a stable entropy [31, 32]. Furthermore, non-linear evolution models already known in statistical physics [33–35] can be easily adapted or generalized within the new theory

 

Finite mixture models deserve further attention in future. A feature of these models is that each of the parameters may be made a function of covariates summarizing household characteristics. Estimation of “heterogeneous” wealth distributions such as these, with distributional shape allowed to vary with personal characteristics, provides a route to decomposition analysis of the sources of differences in wealth inequality across years or countries.20 This could be a good starting point for future research.

 

This point is of particular relevance in the current context, both for the documented presence of long and fat tails towards the upper end of the U.S. net wealth distribution and the fact that all of the three densities accounting for the positive range of wealth obey the weak Pareto law [6]. The weak version of the Pareto law states that the right-hand tail of a distribution behaves in the limit as a simple Pareto model, with an exponent that is a function of the parameters governing the shape of the distribution (see e.g. [58] for an overview). The values of the Pareto index derived from parameter estimates of the Singh-Maddala, Dagum and κ-generalized mixture models are given in the sixth column of Table 2.19 Remarkably, according to the κ-generalized mixture model the set of values for the index of the Pareto tail is closely in the narrow range (1,2] that is generally found in empirical studies on the U.S. wealth distribution [62–64, for instance], whereas for the other two models the Paretian upper ta