12-01-2014 09:54 AM
I have never used boosting sampling, please see the code below! Could anyone explain what this code does and what's the objective of the boosting technique?
I started with train_pot that has a target1=4 and target0=34 and I end up with a train_final that has target1=34 and target0=34 (50:50).
Your help would be much appreciated.
/* Boost the training file to have a 50:50 target1 vs target0 */
proc surveyselect data=train_pot
proc freq data=train_pot.;
table target*numberhits /out=check_boost;
do i = 1 to numberhits;
12-01-2014 08:30 PM
Sorry for late reply,objective is to over sample a rear event in the model development data but using method above you may have multiple same observations in train_final data for target=1 right?