Hello,

I am working with OECD's PIAAC data, a survey of skills of people in OECD countries. The data has 10 plausible values for each skill type. I am using the user-defined -repest- command to analyze the data.

I want to plot the kernel density of the literacy skill (variables pvlit1-pvlit10). But when I use the following command, it only gives me the kdensity plot of the first plausible value pvlit1. I am using the syntax pvlit@, which denotes that pvlit is a plausible value.

Code:
repest PIAAC, est(stata: twoway kdensity pvlit@)
Do you know if its possible to make a kdensity plot that combines all plausible value (not just the kdensity of the first one, as is being done by the code above)? Is it even conceptually sensible to want to combine plausible values or should I be plotting all 10 plausible values separately?

Any help on this would be greatly appreciated!

Thanks.