Hi
i have a question about "copula" package (
http://cran.r-project.org/web/packages/copula/copula.pdf ,
http://www.jstatsoft.org/v21/i04/paper )
Using copulas, I want to generate three random vectors V1, V2 and V3,
where both pairs (V1, V2) and (V1,V3) have the same dependency
structure, defined by copula. To give an example:
I can generate two correlated vectors Va and Vb:
myMvd <- mvdc(copula = gaussianCopula(0.50), margins =
c("norm",
"norm" ), paramMargins = list(list(mean = 0, sd = 2), list(mean = 0,
sd = 2) )
x.samp <- rmvdc(x, 100)
Va = x.samp[,1]
Vb = x.samp[,2]
Doing it again, I will obtain
...
Vc = x.samp[,1]
Vd = x.samp[,2]
What I would like to have is that Va = Vc. After generating Va, I
would like to "fix" it somehow (?), so Vb and Vdare generated with
respect to already generated Va. Is it possible?
--
j
Hi
i have a question about "copula" package (
http://cran.r-project.org/web/packages/copula/copula.pdf ,
http://www.jstatsoft.org/v21/i04/paper )
Using copulas, I want to generate three random vectors V1, V2 and V3,
where both pairs (V1, V2) and (V1,V3) have the same dependency
structure, defined by copula. To give an example:
I can generate two correlated vectors Va and Vb:
? ?myMvd <- mvdc(copula = gaussianCopula(0.50), margins = c("norm",
"norm" ), paramMargins = list(list(mean = 0, sd = 2), list(mean = 0,
sd = 2) )
? ?x.samp <- rmvdc(x, 100)
? ?Va = x.samp[,1]
? ?Vb = x.samp[,2]
Doing it again, I will obtain
? ?...
? ?Vc = x.samp[,1]
? ?Vd = x.samp[,2]
What I would like to have is that Va = Vc. After generating Va, I
would like to "fix" it somehow (?), so Vb and Vdare generated with
respect to already generated Va. Is it possible?
--
j