<s>
Note	O
that	O
Kendall	B-General_Concept
's	I-General_Concept
tau	I-General_Concept
is	O
symmetric	O
in	O
and	O
,	O
whereas	O
Somers’	O
D	O
is	O
asymmetric	O
in	O
and	O
.	O
</s>
<s>
If	O
has	O
a	O
continuous	O
probability	O
distribution	O
,	O
then	O
and	O
Kendall	B-General_Concept
's	I-General_Concept
tau	I-General_Concept
and	O
Somers’	O
D	O
coincide	O
.	O
</s>
<s>
Somers’	O
D	O
normalizes	O
Kendall	B-General_Concept
's	I-General_Concept
tau	I-General_Concept
for	O
possible	O
mass	O
points	O
of	O
variable	O
.	O
</s>
<s>
In	O
practice	O
,	O
Somers	B-General_Concept
 '	I-General_Concept
D	I-General_Concept
is	O
most	O
often	O
used	O
when	O
the	O
dependent	O
variable	O
Y	O
is	O
a	O
binary	O
variable	O
,	O
i.e.	O
</s>
<s>
Methods	O
for	O
fitting	O
such	O
models	O
include	O
logistic	O
and	O
probit	B-Architecture
regression	I-Architecture
.	O
</s>
<s>
Several	O
statistics	O
can	O
be	O
used	O
to	O
quantify	O
the	O
quality	O
of	O
such	O
models	O
:	O
area	O
under	O
the	O
receiver	B-Algorithm
operating	I-Algorithm
characteristic	I-Algorithm
(	O
ROC	O
)	O
curve	O
,	O
Goodman	B-General_Concept
and	I-General_Concept
Kruskal	I-General_Concept
's	I-General_Concept
gamma	I-General_Concept
,	O
Kendall	B-General_Concept
's	I-General_Concept
tau	I-General_Concept
(	O
Tau-a	B-General_Concept
)	O
,	O
Somers’	O
D	O
,	O
etc	O
.	O
</s>
<s>
Identical	O
to	O
the	O
Gini	O
coefficient	O
,	O
Somers’	O
D	O
is	O
related	O
to	O
the	O
area	B-Algorithm
under	I-Algorithm
the	I-Algorithm
receiver	I-Algorithm
operating	I-Algorithm
characteristic	I-Algorithm
curve	I-Algorithm
(	O
AUC	O
)	O
,	O
</s>
