## MRTAnalysis: Primary and Secondary Analyses for Micro-Randomized Trial (MRT) with continuous or binary proximal outcomes

R Package on GitHub R Package on CRAN R Package [...]

R Package on GitHub R Package on CRAN R Package [...]

Contains three functions intended to be helpful in the context of MOST. FactorialPowerPlan calculates sample size needed for factorial experiments under various assumptions. RandomAssignmentGenerator generates a list of random numbers for use as condition numbers to which to assign individuals in a factorial experiment. RelativeCosts1 implements a partly customizable form of a figure in the paper "Design of Experiments with Multiple Independent Variables: A Resource Management Perspective on Complete and Reduced Factorial Designs" (Collins et al., 2009) illustrating the efficiency of factorial experiments.

- Calculate the sample size for a SMART with a longitudinal count outcome, in which the primary aim is to compare the mean count between two adaptive interventions Gallery
#### Calculate the sample size for a SMART with a longitudinal count outcome, in which the primary aim is to compare the mean count between two adaptive interventions

This code calculates the minimum sample size necessary for a SMART with a longitudinal, count outcome, in which the primary aim is a comparison of the mean count between two adaptive interventions starting with different interventions.

This web applet calculates the minimum sample size necessary for a Micro Randomized Trial with continuous proximal outcome.

- Estimate the proximal and distal causal effects of a just-in-time intervention component on a continuous outcome using data from a Micro Randomized Trial Gallery
#### Estimate the proximal and distal causal effects of a just-in-time intervention component on a continuous outcome using data from a Micro Randomized Trial

This code is used to estimate the causal effect of a just-in-time intervention component on the mean of a continuous proximal or distal outcome using data from an MRT.

This code is used to investigate whether (and, if so, how) a baseline or time-varying covariate can be a useful tailoring variables in an adaptive intervention.

Explore

The Center

Opportunities

© 2023 • d3center • Institute for Social Research • University of Michigan