How to select a parameter in fitCopula
Suppose I have two random variables X 1and X 2that are usually distributed. In R, I would generate a sample like this:
> X1 <- rnorm(1000,mean=2.4,sd=1.3)
> X2 <- rnorm(1000,mean=1.5,sd=0.9)
Assuming they are normally distributed with a corresponding mean and sd. My goals are to map the copula C with these specimens, suggesting some family for a bunch of C . For simplicity, we will assume that the bivariate distribution follows the distribution.
The first step is to convert them to a (pseudo) uniform distribution. We will consider U 1= F 1( X 1) and U 2= F 2( X 2sub>). In R, I would do it with the following code
> U1 <- pnorm(X1,mean=2.4,sd=1.3)
> U2 <- pnorm(X2,mean=1.5,sd=0.9)
Then I would put t-copula using the package copula
. I know that I can directly fit the multivariate distribution. However, I would like to know how these things work in a package. A function fitCopula
needs a class object copula
. Obviously I would pass the t-copula. I am not sure how to select the parameter as they should be evaluated. So how can I put t-Copula in U 1and U 2?
source to share
No one has answered this question yet
Check out similar questions: