Відео

Thank you, Professor. My data is not a survey but retrospective data investigating whether some variables (BMI, age, smoking, etc) result in a particular disease outcome. Can I use the survey design function to match propensity scores? Is the MatchIt package the same as the survey design? Thank you for your, sir!

Hello, thank you for your book and this tutorial.I do not have any missing data, also because my data set is not as large. Which steps I can ignore or respectively, in the survey design definition: what should I use instead of "imputation" (because I seem to not need any imputation), if necessary?

Hi Walter. Thanks for the great videos and your excellent book about propensity score matching. I want to estimate PS by your mentioned way; however, I do not understand what these are: ids=~psu, strata=~STRAT_ID, weights=~bystuwt I also read your book, and I can not find the corresponding variables in my dataset. My dataset is the sales data of a company that is not based on a survey. What will happen if I do not include them in the function? I would appreciate it if you could let me know. My other question is that is there ant way to extract propensity scores from the MatchIt package in R? Because matchit() already calculates them for the matching operaion.

Hi! Is there any similar package to do this in Python? Thanks!

Thank you very much. I was using a remote computer of 32GB ram and i left it loading day and night... and it blocked...

Dear Professor Walter, thanks for your kindly presentation and lectures here. I am a medical student in China. I am so happy to see this vedio series here, which helps a lot in my research when I buried myself all day long in a search for a breakthrough in understanding MICE before IPTW and PSM. Thanks! Zhiwen Luo