Hello, I am searching for a way to generate random values from a multivariate mixture distribution. For starters, one could create a mixture distribution consisting of a gamma or normal distribution and then a tail/spike of values in a normal distribution (eg as in Figure 1, A Eyre-Walker, PD Keightley. 2007. The distribution of fitness effects of new mutations. Nature Review Genetics 8: 610-618)

I am not certain how to implement this in the R environment. The implementation becomes even more difficult if one wants to generalize this distribution to multiple variables that are related to one another via some correlation structure. The analogous situation is a multivariate normal distribution with the shape given by the covariance matrix.

What I need for my research program is to generate values for each of 3 random variables with ("one-dimensional") distributions as in Figure 1 and yet simultaneously choose a value for each variable so that the resolved values are correlated in some way (eg if a high value for variable 1 is chosen, it is likely that values 2 and 3 are low).

Thanks in advance for the help.

Patrick SoprovichBiology Department, Dalhousie University

[[alternative HTML version deleted]]

______________________________________________

[hidden email] mailing list -- To UNSUBSCRIBE and more, see

https://stat.ethz.ch/mailman/listinfo/r-helpPLEASE do read the posting guide

http://www.R-project.org/posting-guide.htmland provide commented, minimal, self-contained, reproducible code.