Recently, I wanna reproduce Big Bad Banks’ data processing ,but I find the volume of sample has changed from 5085135 to 5174724.
Could you provide a bit more detail about what you are aiming to reproduce? A link to a specific publication or article is likely the most helpful. It sounds like you are referencing ASEC data between 1977 and 2007, but I am not certain as to your specific question.
https://onlinelibrary.wiley.com/doi/10.1111/j.1540-6261.2010.01589.x Which is the link about the article what I saw. I want to know that if the data about number of household in CPS has been changed between 1977 and 2007 ?Authors said total number of observations in the March Current Population Surveys in the years 1977-2007 is 5,085,135, while I figured out it was 5,174,724.
Thanks for the link. This is helpful. While I can confirm that the sample size of the ASEC samples pooled together from 1977-2007 is 5,174,724, I am not able to find where the authors state their calculation of the sample size in the linked article. Additionally, it seems likely (since IPUMS is not explicitly cited in the article) that the authors used the CPS data available via NBER. These data share the same source, but sometimes handle missing observations differently which can lead to differences in the sample sizes. Alternatively, it could be possible that the authors are conditioning their data on some sort of characteristic (e.g., age, etc.) that makes their total pooled sample size slightly smaller than the full pooled sample.
ok，I’ve got it.Thanks for your help!