seq_mi_fit: MI-based Sequential regression

In Merck/mmrm: Mixed model repeated measures and its extension

Description

Conduct sequential imputation under MAR to get parameter estimates for MMRM.

Usage

seq_mi_fit(seq_formula, data_wide, trt_name, M, fit_model)

Arguments

seq_formula	a list of formulas for sequential regression
data_wide	wide form of the longitudinal data
trt_name	name of treatment in the data frame
M	imputation size
fit_model	type of the analysis model. Available: lm, Rfit::rfit (rank regression), MASS::rlm (robust regression)

Value

A list of estimation results from sequential regression:

• obs_pimarginal probability of the observed data for each treatment.

• beta_ctl_impM sets of the estimated MMRM coefficients used in MI for control group.

• beta_trt_impM sets of the estimated MMRM coefficients used in MI for treatment group.

• sigma_ctlEstimated population covariance matrix of the longitudinal response.

• beta_rubin_ctlEstimated coefficients at the lsat time point for control group.

• beta_rubin_trtEstimated coefficients at the lsat time point for treatment group.

• var_rubin_ctlEstimated covariance matrix of the coefficients at the lsat time point for control group.

• var_rubin_trtEstimated covariance matrix of the coefficients at the lsat time point for treatment group.

Merck/mmrm documentation built on Dec. 17, 2021, 4:12 a.m.