Exercise

Proportion of variance by PCA

PCA results can be visualized in a variety of ways. It is important to choose visualizations that cater to your specific task. For example, we are mainly interested in learning how the components summarize the variation in relation to the overall number of variables. Fewer components means that you are reducing autocorrelations in your data that might be counter to having good predictive models. A good way to explore this is to focus on the proportion of variance explained. A tibble called prop_var has been created for you with the standard deviation from the output list object poke_pca.

Instructions 1/3

undefined XP
    1
    2
    3
  • Create a column pca_comp that enumerates each column in the prop_var table.