Written by: Spencer Greenhalgh
Primary Source: Spencer Greenhalgh
I’m currently working on my dissertation proposal, which has meant exploring principal components analysis. I’ve worked with PCA before, but it’s been a couple of years, so I’m trying to refresh my memory, improve my understanding, and get this proposal moving! Along the way, I’ve found (and been recommended) some helpful resources that I thought I would pass along.
For Understanding/Explaining PCA
My advisor recently pointed me to this CrossValidated (i.e., StackExchange for stats) post that includes an engaging and fun explanation of PCA, complete with a fancy animation. This one isn’t specifically related to R, but it’s a good start for understanding what’s really going on “under the hood” when you carry out a PCA.
For some PCA walkthroughs and R code
R-bloggers recently posted a collection of links to YouTube videos and other resources on using the FactoMineR package in R for doing PCA. I found the videos helpful for getting back up to speed with PCA after a couple of years away, and I’m going to keep FactoMineR in mind in the future.
For plotting PCA results
I’m interested in graphical displays of PCA results and of using visualizations to interpret PCA; while I was looking for help here, I found an answer on Cross-Validated that not only provided some helpful advice but also made available some code for creating nice plots of PCA data.
I’m sure I’ll stumble upon some other helpful resources as I continue to work with PCA and move my dissertation forward!