author's proof - DARP
author's proof - DARP
author's proof - DARP
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
AUTHOR'S PROOF<br />
Reference distributions and inequality measurement<br />
JrnlID 10888_ArtID 9238_Proof# 1 - 08/01/13<br />
Q2<br />
Fig. 2 Quantile approach with<br />
the most equal reference<br />
distribution<br />
x<br />
Q3<br />
°<br />
°<br />
°<br />
°<br />
Δ x 3<br />
°<br />
x 3<br />
F −1 *<br />
(·)<br />
1<br />
2.2 Reference distributions 96<br />
The use of the J index also requires the specification of a reference distribution: there 97<br />
are several possibilities. 98<br />
The most equal reference distribution Let us assume that the most equal income 99<br />
distribution is when the same amount is given to each individuals:<br />
100 Q4<br />
( ) i<br />
F∗<br />
−1 =ˆμ for i = 1,...,n (4)<br />
n + 1<br />
If we use this (egalitarian) distribution as the reference distribution in Eq. 3,thenwe 101<br />
find the standard Generalised Entropy inequality measure: 8 102<br />
1<br />
n∑<br />
[( ) α xi<br />
J α =<br />
− 1]<br />
,α̸= 0, 1 (5)<br />
nα(α − 1) ˆμ<br />
UNCORRECTED PROOF<br />
i=1<br />
So the GE inequality measures are divergence measures between the EDF and the 103<br />
most equal distribution, where everybody gets the same income. They tell us how far 104<br />
a distribution is from the most equal distribution. A sample with a smaller index has 105<br />
amoreequal distribution. 106<br />
Figure 2 presents the quantile approach for this case. We can see that the EDF is 107<br />
always above (below) the reference distribution for large (small) values of incomes. 108<br />
It makes clear that large (small) values of α would be more sensitive to changes in 109<br />
high (small) incomes. 110<br />
8 The limiting forms are J 0 =−n 1 ∑ ( )<br />
ni=1<br />
log xi<br />
, J ˆμ 1 = n 1 ∑ ( )<br />
ni=1 x i<br />
ˆμ log xi<br />
. ˆμ