How to “Use PCA or PARAFAC to investigate outliers”

If you decide to perform a PCA model
you’ll find that most standard procedures such as the MATLAB built-in SVD do not handle missing data.

There are at least two ways to circumvent this
All standard chemometric software such as Unscrambler, SIMCA, PLS_Toolbox etc.
have (sub-optimal) algorithms for handling missing data but you can also find dedicated algorithms that use e.g. imputation to handle the missing data more appropriately.

You may get MATLAB code for this
from Rasmus Bro.