Parameter estimation for multivariate skew-normals
Unlike multivariate normal, where the parameters estimates can be obtained using the sample mean and sample variance-covariance matrix, the parameters of the skew-normal distribution need to be estimated by an iterative process. You can make use of existing functions in the sn packages to get the estimates.
In the previous exercise, we examined the Ht and Wt variables from the ais.female object and determined that a skew-normal distribution can model the distribution. Now, we will estimate the parameters of this distribution.
This exercise is part of the course
Multivariate Probability Distributions in R
Exercise instructions
- Calculate the parameters of the skew-normal distribution and assign it to object
fit.ais. - Print only the skewness parameters.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Fit skew-normal parameters
fit.ais <- ___(y = cbind(ais.female$Ht, ais.female$Wt), opt.method = ___)
# Print the skewness parameters
fit.ais$dp$___