Dear Statalist members,

I am currently working with a household survey panel dataset to create a financial vulnerability index (FVI) for households for the year 2014 to 2020. The survey data contains information on various variables that can be used to define the financial vulnerability of the households. I will be using approximately 10 variables to create the index. The variables are mostly categorical or dichotomous except for one variable which measures the income from sale of assets. The categorical and dichotomous variables include: If the household faced issues in making payments (1=yes, 0=no), How is the household fairing financially as compared to last year (1=worse, 2=the same and 3=worse), and so on.
The literature on creating similar indices usually use nonlinear PCA method (NLPCA) owing to the nature of the data which is a generalised version of the PCA methodology. I have read about conducting PCA analysis in Stata using the pca command. However, I was unable to find help on creating an index using NLPCA in Stata. It would be really helpful if someone could help me with the command for using NLPCA to create the FVI? Any further help on NLPCA methodology in general would also be appreciated.
Since most of the variables are dichotomous or categorical in this case, would you suggest using NLPCA? or if there is other methodologies more suitable for creating this index?

Any help will be much appreciated!

Thank you.

Best,
Akshita