matrix with mixed data ?

I have some variables of mixed type (categorical and numerical)

I would like to build a table or cell array (or a matrix in general) in which all of them are listed, one per column.

categorical values can be repeated for certain row ranges and can be equal, numerical data are individually different.

what type of matrix would be better to perform statistics ?

a table ? a cell array?

thanks !

Answer 1

[Jos (10584)]

A table is the most obvious choice for this, especially if you want to do statistics. I suggest you read the manual on tables.

[LO]

my plan is to perform CATPCA (categorical principal component analysis). would that work ?

thanks !

[ErikaZ]

LO, were you able to run CATPCA in Matlab?

[LO]

No, but I ended up using a package for R-studio (Factominer) which was quite handy and easy to use. I would recommend trying: http://www.sthda.com/english/articles/31-principal-component-methods-in-r-practical-guide/115-famd-factor-analysis-of-mixed-data-in-r-essentials/