About this document | Parallel computing setup | Simulated data generating process | Generate datasets | Model fit | 1. Linear mediator model, linear outcome model | 2. Logistic mediator model, linear outcome model | 3. Linear mediator model, logistic outcome model | 4. Logistic mediator model, logistic outcome model | Results comparison

No EMM by covariates | EMM by covariates | There is $A\times C$ term in mediator model | There is $A\times C$ term in both mediator and outcome models | There are $A\times C$ term in both mediator and outcome models, and $M\times C$ term in outcome model

Data preparation | regmedint object | regemedint() object constructor | summary() for regmedint | coef() for regmedint | vcov() for regmedint | confint() for regmedint | print() for regmedint | Methods for summary_regmedint | coef() for summary_regmedint | print() for summary_regmedint

mreg linear yreg linear (V2015 p466 Proposition 2.3) | Point estimates | Standard error estimates | mreg linear yreg non-linear (V2015 p468 Proposition 2.4) | mreg logistic yreg linear (V2015 p471 Proposition 2.5) | mreg logistic yreg non-linear (V2015 p473 Proposition 2.6) | Bibliography

Bootstrapping support | boot package | modelr package | Comparison to the delta method confidence intervals

Missing data generation | Truth fit | Naive complete case analysis | Multiple imputation | Comparison