[Date Index] [Thread Index] [Author Index]

Re: How to sample a 2-dim. r.v. with known density function?

• To: mathgroup at smc.vnet.net
• Subject: [mg65296] Re: How to sample a 2-dim. r.v. with known density function?
• From: "KvS" <keesvanschaik at gmail.com>
• Date: Fri, 24 Mar 2006 01:00:05 -0500 (EST)
• References: <dvu47b$86p$1@smc.vnet.net>
• Sender: owner-wri-mathgroup at wolfram.com

Hi Tony,

thanks for your informative response! I looked up some stuff on the
rejection method you mention and I was able with some basic
transformation and estimation to find a dominating density that
represents just two independent Gaussians conditioned to be
non-negative :). I understand that the ratio p(x,y)/p0(x,y) is
important in order to avoid getting too many rejections before
acceptance, I have to yet to try and see if the estimate is good enough
in this sense.

- Kees