I am curious if there is an easy way to select a proportional sample out oft a panel data that …
- … is restricted to a specific number of observations (i.e. sample = 100) in the new sample
- … comes up with the best sample fit (in terms of mean, variance etc.) compared to the population (panel data)
- … considers (calculates and selects the new sample based on) more than five different measurements/variables
It should be still very similar to the original population in the panel data.
Is there anyone that could help me with that?
Thank you.
Konstantin
0 Response to Proportional sampling based on population mean, variance etc.
Post a Comment