Generate samples from multivariate t-distribution
Though multivariate normal is widely used, not all multivariate data follow a normal distribution. Multivariate t-distributions can accommodate heavy-tailed distribution in each direction. In this exercise, you will learn how to draw random samples from a multivariate t-distribution. We will use the same mu.sim and sigma.sim parameters that were used to generate samples from multivariate normal distributions.
Cet exercice fait partie du cours
Multivariate Probability Distributions in R
Instructions
- Generate 200 samples from a bivariate t-distribution with 5 degrees of freedom. Label the object
multt.sample. - Print the first six samples.
- Check whether the samples follow a multivariate normal distribution using the Mardia Test and plot the relevant qqplot for the test.
Exercice interactif pratique
Essayez cet exercice en complétant cet exemple de code.
# Generate the t-samples
multt.sample <- ___
# Print the first 6 samples
# Check multivariate normality
mvn(___, mvnTest = "___", multivariatePlot = "___")