The stata manual for collapse (p. 7) states that using aweight when calculating the sum with collapse normalizes the weights to sum to number of observations in a group. Can someone please explain what does that exactly mean? How does the exact formula look like for the weights when I calculate the sum in cell? And when should I use what weights (aweight vs f/i/pweight)?
Related Posts with Collapse Weight Normalization
Power for one group proportion where sample size is sufficient for lower 95%CI to be above a certain threshold- CI for smoothed scaled Schoenfeld residuals after fitting Cox model
- Census division by year fixed effects
- Endogeneity with Hausman test?
- margins - factor variables may not contain noninteger values
Subscribe to:
Post Comments (Atom)
0 Response to Collapse Weight Normalization
Post a Comment