Hello Everyone,
I would like to achieve one function in R and not sure if there is an existing function in the community so that I don't need to program it by myself. Basically, as you can see in the attached image, I have several people with different personID (3, 4 and 5 in the image). There are multiple roles for every personID with a different weightPercent. I want R to do a random selection based on the weightPercent so that only one role will remain. For instance, for personID 4, I only want one role of data left after R did the selection, and lines 10, 11, 12 and 13 have 36.8%, 21.0%, 21.0% and 21.0% of probability to be selected respectively.
I tried to use the formula sample_n like this:
a <- sample_n(demobel_matched, 1, weight = demobel_matched$weightPercent, replace = TRUE)
However, this does not return the thing I want and I got an error saying "incorrect number of probabilities" in this formula. So I think sample_n may not achieve the function I want.
Hence, I am wondering if you have any suggestions or existing known formulas to achieve my goal? Thanks very much for your kind help in advance!
