I am trying to understand how to derive the quantile function from the cdf.
The Singh-Maddala cdf is
Should i just solve for x ?
Thank you
$\begingroup$
$\endgroup$
2
-
2$\begingroup$ Yes: that's what "deriving" the quantile function means, doesn't it? $\endgroup$– whuber ♦Commented Aug 2, 2016 at 23:05
-
$\begingroup$ This may help you see that the answer is "yes". $\endgroup$– Glen_bCommented Aug 3, 2016 at 10:19
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
In case you were still wondering, solving $F(x)=u$ in $x$ leads to $$ Q(u) = b\Big((1-u)^{-1/q}-1\Big)^{1/a},$$ where $u$ is a realisation of standard uniform variate. To generate Singh-Maddala random variates, $1-u$ can be replaced simply by $u$.
