http://www.sthda.com/english/articles/31-principal-component-methods-in-r-practical-guide/112-pca-principal-component-analysis-essentials/

University of Michigan - Ann Arbor

The principal components represent the directions of the data that captures the maximum variance (in other words, most information). More specifically, principal components are the individual dimensions that are found to be "interesting" through PCA.

What are Principal Components?

You can use a correlation circle to show the principal components. These visualizations show the relationship between the original datapoint and the new principal components. On the graph below, the vertical line through the center of the circle represents the first principal component, which we can see explains 41.2% of variation in the data. The labeled arrows that are more perpendicular to the first principal component are explained most accurately by the first component. For example, the X400m, X110m hurdle, and X100m are close to each other and close to perpendicular to the first principal component line meaning that these three variable are likely related and can be best explained by the first principal component.



Learn Before

Related