Let $X$ and $Y$ be non negative random variables with joint distribution \begin{equation} F(x,y)=1-e^{-x}-e^{-y}+e^{-x-y-\delta xy}; ~~~x\geq 0,~~y\geq 0. \end{equation} How to generate a bivariate random sample of size $n$ for different values of $\delta$ associated with the same distribution.
Asked
Viewed
80 times
1
$\begingroup$
$\endgroup$
-
2$\begingroup$ check with copula generation $\endgroup$ – Xi'an May 28 '15 at 10:58
-
1$\begingroup$ This question looks like a special case of the general problem addressed in (the near-duplicate) threads at stats.stackexchange.com/questions/124865, stats.stackexchange.com/questions/133881, and stats.stackexchange.com/questions/149026. If they do not solve your problem, would you please edit this post to indicate where specifically you need help? $\endgroup$ – whuber♦ May 28 '15 at 13:22
-
1$\begingroup$ See pp. 583-584 in chapter 11 of Luc Devroye's book Non-Uniform Random Variate Generation $\endgroup$ – Yves May 28 '15 at 13:28
