[Elif Cengen]

Dear Stephen,

Great news! Many thanks. And it is also great that results are saved in r().

[Elif Cengen]

Dear Stephen,

Great news! Many thanks. And it is also great that results are saved in r().

[Elif Cengen]

Dear Stephen, great news! Many thanks!

[Elif Cengen]

Dear Stephen, great news! Many thanks

[shem shen]

ineqdecgini is now available from SSC

Thank you for making this available to us!
By the way, is there any user-written command in Stata that allows us to compute the economic distance ratio (varies between 0 and 1) between two population subgroups, as defined in Dagum, C. (1980) "Inequality Measures Between Income Distributions With Applications" Econometrica 48 (7), 1791-1803. ?

[Stephen Jenkins]

The more important question is why you'd want to use such a measure in the first place. Dagum's paper has been severely criticized. See e.g.

On the Distance between Income Distributions
Author(s): Anthony F. Shorrocks
Source: Econometrica, Vol. 50, No. 5 (Sep., 1982), pp. 1337-1339
Published by: The Econometric Society
Stable URL: http://www.jstor.org/stable/1911879

[shem shen]

The more important question is why you'd want to use such a measure in the first place. Dagum's paper has been severely criticized. See e.g.

Thank you so much Professor Jenkins! I am not aware of this critique. I will read this paper.
I am interested in this measure because I am interested in decomposing the Gini index into a "clean" between- and within-group component (with no residual term) according to the weakly decomposability principle (Pauline Mornet, 2013. "A program for weakly decomposable inequality measures by population subgroups," Economics Bulletin, AccessEcon, vol. 33(3), pages 1738-1750. http://www.accessecon.com/Pubs/EB/20...33-I3-P163.pdf). On page 1741 the author shows how the "gross" between-group component of Gini can be written as a function that contains the economic distance ratio as a scalar.

See Stéphane Mussard & Michel Terraza & Françoise Seyte, 2003. "Decomposition of Gini and the generalized entropy inequality measures," Economics Bulletin, AccessEcon, vol. 4(7), pages 1-6. for a very similar exposition.

I am not sure if the author is reputable enough in this field but I also see a recent application of this type of "weak decomposition of gini" in Modalsli, J. (2017). Decomposing global inequality. Review of Income and Wealth, 63(3), 445-463. https://onlinelibrary.wiley.com/doi/...111/roiw.12230 .

[Stephen Jenkins]

Personally, I am not very attracted to the decompositions presented in those references. If you want a "clean" decomposition such that total inequality = Within-group inequality (weighted sum of subgroup inequalities) plus Between-group inequality, then use of members of the generalised entropy family is the only way to proceed. Gini decompositions of this kind inevitably end up with a residual term (except in special cases). As I pointed out earlier in this thread, some researchers argue that the residual term can be usefully interpreted in terms of 'overlapping'.

[shem shen]

Personally, I am not very attracted to the decompositions presented in those references. If you want a "clean" decomposition such that total inequality = Within-group inequality (weighted sum of subgroup inequalities) plus Between-group inequality, then use of members of the generalised entropy family is the only way to proceed. Gini decompositions of this kind inevitably end up with a residual term (except in special cases). As I pointed out earlier in this thread, some researchers argue that the residual term can be usefully interpreted in terms of 'overlapping'.

Thank you!