[R] error message fitting tcopula

From: Roslina Zakaria <zroslina_at_yahoo.com>
Date: Wed, 09 Jun 2010 23:36:38 -0700 (PDT)


Hi r-users,
 
I really need help in fitting the t-copula.  I try to reproduce the example given by Jun Yan in “Enjoy the joy of copula” but I’m not sure how to correct the error based on the error message.  I tried so many ways but still could not get it working.  
loglik.marg <- function(b, x) sum(dgamma(x, shape = b[1], scale = b[2], log = TRUE))  
ctrl <- list(fnscale = -1)
 
#dat <- stn_pos[,1:2] ## observed data
myCop.t <- ellipCopula(family = "t", param = 0.5,dim = 2, dispstr = "un", df = 8) myCop.t
 
myMvd <- mvdc(copula = myCop.t, margins = c("gamma", "gamma"), paramMargins = list(list(shape = 1.5, scale = 38), list(shape = 1.7, scale = 50)))
myMvd
n <- 200
dat <- rmvdc(myMvd, n)
 
mm <- apply(dat, 2, mean)
vv <- apply(dat, 2, var)
rho <- rcorr(dat,type="spearman")[[1]]; round(rho,2) rbind(mm,vv)
 

b1.0 <- c(mm[1]^2/vv[1], vv[1]/mm[1])
b2.0 <- c(mm[2]^2/vv[2], vv[2]/mm[2])
a.0  <- sin(cor(dat[, 1], dat[, 2], method = "kendall") * pi/2)
start <- c(b1.0, b2.0, a.0)
 
b1hat <- optim(c(1,5), fn = loglik.marg, x = dat[, 1], control = ctrl)$par  
b2hat <- optim(c(1,5), fn = loglik.marg, x = dat[, 2], control = ctrl)$par  
udat <- cbind(pgamma(dat[, 1], shape = b1hat[1], scale = b1hat[2]),pgamma(dat[, 2], shape = b2hat[1], scale = b2hat[2]))  
> fit.ifl <- fitCopula(myCop.t, udat,  method="mpl", c(1,5,1,5,8),estimate.variance=TRUE)
Error in fitCopula.ml(data, copula, start, lower, upper, optim.control,  :   The length of start and copula parameters do not match.  
> fit.ifl <- fitCopula(udat, myMvd_at_copula, c(1,5,1,5,8))
Error in fitCopula(udat, myMvd_at_copula, c(1, 5, 1, 5, 8)) :   Implemented methods are: ml, mpl, itau, and irho. In addition: Warning messages:
1: In if (method == "ml") fit <- fitCopula.ml(data, copula, start,  :   the condition has length > 1 and only the first element will be used 2: In if (method == "mpl") fit <- fitCopula.mpl(copula, data, start,  :   the condition has length > 1 and only the first element will be used 3: In if (method == "itau") fit <- fitCopula.itau(copula, data, estimate.variance) else if (method ==  :   the condition has length > 1 and only the first element will be used 4: In if (method == "irho") fit <- fitCopula.irho(copula, data, estimate.variance) else stop("Implemented methods are: ml, mpl, itau, and irho.") :   the condition has length > 1 and only the first element will be used  
> fit.ifl <- fitCopula(udat, myMvd_at_copula, a.0)  ## this given the paper
Error in fitCopula(udat, myMvd_at_copula, a.0) :   Implemented methods are: ml, mpl, itau, and irho.  
Thank you so much for your help.
      
	[[alternative HTML version deleted]]


______________________________________________

R-help_at_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. Received on Thu 10 Jun 2010 - 06:39:28 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Thu 10 Jun 2010 - 06:40:29 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive