The package FactoInvestigate
The package FactoInvestigate describes and interprets automatically the results of your principal component analysis (PCA, CA or MCA), choosing the best graphs to show.
Just just have to do your analysis as usual and then use Investigate(res) to obtain the following report on the results of your principal component analysis.
Why using FactoInvestigate?
- It proposes automatically an first interpretation of your results
- It does the best selection of the labels of individuals and variables to have the most readable graphs
- It selects the number of dimensions that should be interpreted
- It detects the potential outliers and redo the analysis after analysing the specificities of the outliers
- The "best" (supplementary) qualitative variable is used to colour the individuals according to that variable
- A clustering is performed on the principal component results and then interpreted
- The report is available in pdf, Word or html but also with an RmarkDown file
- You can modify the default parameters for the selection of your analysis
- The report detects the language of your R session and write the report in this language (French or English)
How to use FactoInvestigate?
You simply have to perform your analysis with FactoMineR, as usual, and then to use the function Investigate of the package FactoInvestigate.
library(FactoMineR)
data(decathlon)
res = PCA(decathlon, quanti.sup = 11:12, quali.sup=13, graph=FALSE)
Investigate(res)
Then you will obtain this report automatically in html (click to enlarge):
