Multidimensional scaling in three dimensions
In this exercise, you will perform multidimensional scaling of all numeric columns of the wine data, specifying three dimensions for the final representation.
This exercise is part of the course
Multivariate Probability Distributions in R
Exercise instructions
- Exclude the first column containing the wine type and compute the distance matrix for all other columns and assign it to an object
wine.dist. - Compute the multidimensional scaling of the data to represent the distance in three dimensions.
- Use the
scatterplot3d()function to plot the representation and color them by wine type.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Calculate distance
wine.dist <- dist(___[,-1])
# Perform multidimensional scaling
mds.wine <- cmdscale(___)
mds.wine_df <- data.frame(mds.wine)
# Plot the representation of the data in three dimensions
scatterplot3d(___, color = ___, pch = 19, type = "h", lty.hplot = 2)