Drawing Samples based on Frequency Weights of another Group
Hello Everyone!
I have been trying to figure out a problem and would appreciate any help with the same:
I have two groups (A and A-residual) which are put in strata groups.
Group A has 2000 unique observations that I want to use for the experiment. Group A-residual has 1000 unique observations. There are 120 strata groups. I want to randomly draw another 500 observations from Group A-residual in the same proportion as the strata groups in Group A.
E.g.:
If there are 200 observations in Strata 1 in Group A then the proportion of Strata 1 in Group A is 200/2000=0.1
This would mean that in the case of Group A residual, the observations to be drawn are 200*500/2000=50 (the same proportion as strata 1 in group A - 50/500=0.1)
It would be great if I could get any ideas as to how I can go about the same.
Thanks a lot!
I have been trying to figure out a problem and would appreciate any help with the same:
I have two groups (A and A-residual) which are put in strata groups.
Group A has 2000 unique observations that I want to use for the experiment. Group A-residual has 1000 unique observations. There are 120 strata groups. I want to randomly draw another 500 observations from Group A-residual in the same proportion as the strata groups in Group A.
E.g.:
If there are 200 observations in Strata 1 in Group A then the proportion of Strata 1 in Group A is 200/2000=0.1
This would mean that in the case of Group A residual, the observations to be drawn are 200*500/2000=50 (the same proportion as strata 1 in group A - 50/500=0.1)
It would be great if I could get any ideas as to how I can go about the same.
Thanks a lot!
