L8: Potential outcomes and IPW

Readings

• Cole SR, Hernán MA. Adjusted survival curves with inverse probability weights. Computer methods and programs in biomedicine. 2004;75(1):45–9.

Slides

Download PDF of all slides

Notes

This week’s reading is an excellent introduction to IPW for survival functions. As you read, be sure you can answer the following questions:

• Why was the paper written? What problem was it trying to solve?
• What is the point of the first paragraph on page 46?
• Why are non-parametric standardization and IPW infeasible in the presence of multiple or continuous covariates?
• On page 46, what do they mean by “prior logistic regression models?”
• Why use stabilized vs unstabilized weights? What is the difference?
• What is the “pseudopopulation?”
• What does the “+” mean in Table 1?
• What is the point of Table 2?

Videos

Recordings of lectures are available on the EPID 722 Sakai site.

Exercise

Exercise 8 may be found on the course Teams site.

Questions

Please use the form below to submit your questions about this week’s reading.