Standardization for time-fixed treatments

Learning objectives

1. Describe standardization to estimate marginal effects.
2. Interpret standardized estimates
3. Use modeling to estimate standardized estimates with many covariates.
4. Describe bootstrapping to obtain 95% confidence intervals.
   
   

Notes

The R code uses data from the NHANES I Epidemiologic Follow-Up Study (NHEFS). You can download the data used for this code here.

Additional readings or resources

1. Causal Inference: What If (Chapter 13)
2. Ahern J, Hubbard A, Galea S. Estimating the effects of potential public health interventions on population disease burden: a step-by-step illustration of causal inference methods. American journal of epidemiology. 2009 May 1;169(9):1140-7.