I have dataset only with number of adults and their incomes, and total health OOP expenditures for whole household. I used IC2 package in Rstudio to calculate concentration index for a income(C), and gini index for a OOP (G). I calculated it without weights and with default parametrs. Kakwani = C-G.
Dear Askhat Shaltynov ,
If you have a survey with sampling weights, typically, probabilist weights, you should use them in order to get estimates that are representative for the whole population.
In Stata, a very good alternatives are the DASP package (http://dasp.ecn.ulaval.ca/) or conindex (https://doi.org/10.1177%2F1536867X1601600112).
In R, which I do not use very much, I see that the packages rineq (https://rdrr.io/github/brechtdv/rineq/) or ic2 (https://rdrr.io/github/brechtdv/rineq/) allow this sort of calculation using weights.
I hope it helps.
Best,
José-Ignacio
Dear José!
Thanks a lot and Happy New year!
Unfortunately we have not a datasets with weights of households. Is it possible to calculate the
Kakwani index and catastrophic health expenditures without weights?
Yes, of course.
The weights of the surveys are present when there is not a simple random sampling (among others). I meant that you should weights if they are provided.
- Badges
- Science topic
- Similar topics
- Biological Science
- Zoology
- Ornithology
- birds
- Scopus