[Stephen ONeill]

Hi Hoang,
I am glad you are finding it helpful. For a binary variable, the bounds for the index are -(1-μ ) and (1-μ ) since W = CI / (1-μ ).
For a non-binary bounded variable, bounded between a and b: W = (μ(b − a) / [(b − μ)(μ − a)]) CI

and W is bounded between - μ(b − a) / [(b − μ)(μ − a)] and μ(b − a) / [(b − μ)(μ − a)]

Suppose the variable H was bounded between 1 and 3 and the ranking variable was income we would write:
conindex H, rankvar(Income) limits(1 3) bounded wagstaff

At the risk of self-promotion, our paper https://www.stata-journal.com/articl...article=st0427 discusses the various indices and what sort of inequality they measure. Feel free to email me directly at [email protected] if you have further questions.

Regards,
Stephen

[Hassen Ali]

Thank you very much!! I have learned a lot from your posts
Cheers ,Hassen

[ankiti arya]

I wonder if I can draw a graph of Erreygers Corrected Concentration Index. However, I do not know any command of that, If anyone have the command please do share it.

Thanks

[Yuan Hui]

Hi Stephen,
I have some troubles when using the "conindex" command. Each row of my dataset represents the risk suffered by a group of people. I use the following command to draw concentrated curves: conindex newrisk, truezero rankvar(lowincom_rate) graph or conindex newrisk [aweight=Population], truezero rankvar(lowincom_rate) graph. However, it seems that the results is not right, because the "conindex" command seems to regard each row of the dataset as the information of one person. So the abscissa of concentration curve seems not be sorted by population percentage(As shown in the figure below).

The dataset format is as follows:

What could I do? Could you please help me solve this problem. Thanks a lot.
Yuan Hui

[Stephen ONeill]

Hi Yuan Hui,
I wonder if you could share the data with me - [email protected] and I will have a look?
Regards,
Stephen