After propensity score matching if we are getting sample size of group A around 20 times of other group B. Baseline characteristics are properly balanced after PS matching. But what about the conclusions drawn on this kind of unbalanced data (in terms of sample size)?
Na = 680000 and Nb=36000. Is there any guidance or literature available for this?
I think you will find right answer from the following study
Decreasing the sample size from 1,000 to 40 subjects did not substantially alter the Type I error rate, and led to relative biases below 10%. The IPTW method performed better than the PS-matching down to 60 subjects. When N was set at 40, the PS matching estimators were either similarly or even less biased than the IPTW estimators. Including variables unrelated to the exposure but related to the outcome in the PS model decreased the bias and the variance as compared to models omitting such variables. Excluding the true confounder from the PS model resulted, whatever the method used, in a significantly biased estimation of treatment effect. These results were illustrated in a real dataset.
Article Evaluation of the Propensity score methods for estimating ma...
If you prefer to have a balanced sample size such as 1:1 matching, you can use proc psmatch in SAS software. Be sure to include k=1 in the match statement. You can also vary your k=x to allow for 1:x matching. Find link below:
https://support.sas.com/documentation/onlinedoc/stat/142/psmatch.pdf
1 - First point - group A not around 20 times of other group B because if its like that than you get 720000 in group A but you are getting 680000. that means you just give command in the software , and not give direction how much you want matching group like 1:1 matching and 1:2 matching .
2nd point , if you just give command than its give you m to 1 matching which also one type of propensity score matching .. for this , that article will help you.Article Statistical Criteria for Selecting the Optimal Number of Unt...
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size