08-11-2015 11:35 AM
I tried to do a Regression Analysis (with backward elimination) for my data set and it worked fine. However, when I tried to use Data Partition to split the data into Training (60%) and Validation (40%), it comes with the error
" The computed Average Squared Error for Ln_NW is 6.699147E-24.
This indicates a possible target duplication issue.
Please review the list of inputs used in the model."
08-14-2015 10:41 AM
This message indicates that since the model is predicting almost perfectly (near-0 ASE), that there is likely an input variable that has identical or almost identical values as the target variable. However, it is very strange that you didn't get that message when running with the un-partitioned data. In the Results from the un-partitioned Regression, are you seeing a very small ASE?