Thanks a lot! It works, however it is less smooth than I had expected. And I have a further question: What if I have my own 2-d density function as: f <- function(x1,x2,parameters) ... and there is no way how to generate random variates (x1,x2) from f. But I'd still like to get the contour densities plots for (x1,x2) at levels 10^(-j/2) for j=2,...,9. Is there a method to do that? A smooth plot would be even more useful in my case. LX
ÔÚ2008-10-04 22:42:00£¬"Eik Vettorazzi" <[EMAIL PROTECTED]> дµÀ£º >Sorry, mixed up where to place the desired level, it has to be done in >the plot part, as in > >d <- kde2d(z1,z2) >contour(d, levels=10^(-(2:9)/2), add=T) > > >Eik Vettorazzi schrieb: >> You may use kde2d from MASS, with estimates a 2d density >> >> d <- kde2d(z1,z2,h=10^(-(2:9)/2)) >> plot(z1,z2) >> contour(d,add=T) >> >> hth. >> >> biyeshejiqx schrieb: >>> Hello,everybody, >>> >>> I used the following codes to generate bivariate normal dependence >>> structure with unit Frechet margins. Sigma <- >>> matrix(c(1,.5*sqrt(1),.5*sqrt(1),1),2,2) # generate y <- >>> mvrnorm(Nsam, c(0,0), Sigma) # random v <- cbind(pnorm(y[,1],mean = >>> 0, sd = 1), pnorm(y[,2],mean = 0, sd = 1)) >>> z <- cbind(-1/log(v[,1]),-1/log(v[,2])) >>> z1 <- z[,1] >>> z2 <- z[,2] >>> And to get the scatter plot by: plot(z1,z2) >>> >>> How can I get the contour densities plots for (z1,z2) at 10^(-j/2) >>> for j=2,...,9? >>> >>> Waiting for your reply!Many thanks! >>> >>> Xiao >>> >>> >>> [[alternative HTML version deleted]] >>> >>> ______________________________________________ >>> R-help@r-project.org mailing list >>> https://stat.ethz.ch/mailman/listinfo/r-help >>> PLEASE do read the posting guide >>> http://www.R-project.org/posting-guide.html >>> and provide commented, minimal, self-contained, reproducible code. >>> >> >> ______________________________________________ >> R-help@r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. > [[alternative HTML version deleted]]
______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
