turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- General Programming
- /
- Atkinson coefficient

Topic Options

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-18-2016 09:32 AM

Hello,

I need help with programming the Atkinson Coefficient for in-eqaulity.

I couldn't find anything helpful in the internet, i have data which contains id number and income of each people.

I want the program to calculate the Atkinson coefficient, does somebody has a code of this?

Thank you for helping,

Matan.

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Greisas

04-18-2016 11:00 AM

If you have IML you do nto need to use my macro

https://goo.gl/cfBKUc

https://communities.sas.com/t5/General-SAS-Programming/Atkinson-coefficient/m-p/264557

%utl_submit_r64(%nrbquote(

library(ineq);

/* generate vector (of incomes) */

x <- c(541, 1463, 2445, 3438, 4437, 5401, 6392, 8304, 11904, 22261);

/* compute Gini coefficient */

ineq(x);

/* compute Atkinson coefficient with parameter=0.5 */

ineq(x, parameter=0.5, type="Atkinson")

));

[1] 0.4620911

[1] 0.1796591

In response to another post the Whittaker W/M functions are also in R packages

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to rogerjdeangelis

04-19-2016 02:25 AM

thanks for the help,

but what you gave me is a package for R, and i need to write the code in SAS.

also i don't use the SAS Enterprise Guide but the basic SAS.

thanks.

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Greisas

04-18-2016 11:27 AM

Hi,

Atkinson's inequality measure can be calculated using the formula, say

(yi/ybar)^.5 Atkinson, e=0.5

(yi)^(1/n) Atkinson, e=1

(yi/ybar)^(-1) Atkinson, e=2

where yi is the income of the each people,

ybar is the mean value of the income of all.

n is the number of people.

e is the weighting parameter.

Regards,

Manohar

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Manu_SAS

04-19-2016 02:29 AM

Thank you,

So if i understand correctly, for atkinson with parameter 0.5, I don't need to use the number of people in my data set.

Also, if I want paramter 2.5, what formula do I need?

So if i understand correctly, for atkinson with parameter 0.5, I don't need to use the number of people in my data set.

Also, if I want paramter 2.5, what formula do I need?

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Greisas

04-19-2016 03:24 AM

Please refer the heading **6.2.3 Atkinson's inequality measures** in the attached pdf.

This is might be helpful.

Replying with formula is quite diffcult.

Apply the value of weighting parameter with the formulae gives you the result.

Regards,

Manohar