About this code
This package includes a series of functions, listed below are the main functioning ones:
- parameter_adjust helps to adjust initial parameters to achieve certain effect size (signal-to-noise ratio) of main effect and covaraite effect.
- generate_SMART generates clustered SMART data based on given parameters for simulation testing.
- solve_SMART does all analysis for clustered SMART data, including effect estimate, working variance model estimate, as well as variance estimator.
- hypothesis_testing does more generalized hypothesis testing using solve_SMART output.
- data_expand duplicates data to make it capable of being analysed in off-the-shelf geepack.
How can a behavioral scientist use this code?
A behavioral scientist can use this code to learn how to analyze Clustered SMART data.