"Warning: no valid values" using Synthesize

9 views
Skip to first unread message

Frank Fineis

unread,
Apr 25, 2016, 4:40:32 PM4/25/16
to IVEware
I've been using R to write .txt files to the datasets/datain directory, and from there proceeding through the normal steps to create synthetic datasets.

Upon hitting "synthesize," the job immediately fails with a log file with the following warnings:

Warning: No valid values for StudentGender13
           Type changed to transfer
Warning: No valid values for ReadingLevel13
           Type changed to transfer
Warning: No valid values for Race13
           Type changed to transfer
...(more warnings for ~8 other categorical variables)

And then failure was due to "Error: Imputation too big - possibly continuous variable specified as categorical," which is understandable I suppose given I have two categorical variables with ~1000 different levels.

My main concern is regarding the warning. I've tried importing data as both quoted and then unquoted; is this easily diagnosed?


Thank you,
Frank
Reply all
Reply to author
Forward
0 new messages