stan.logisticmodeltrans {BayesianPlatformDesignTimeTrend} | R Documentation |
This function transform the data in trial simulation to the data required for stan modelling
stan.logisticmodeltrans(
z,
y,
randomprob,
group_indicator,
armleft,
group,
variable.inf,
reg.inf
)
z |
A vector of all treatment index data from the beginning of a trial |
y |
A vector of all outcome data from the beginning of a trial |
randomprob |
A named vector of randomisation probability to each arm |
group_indicator |
A vector for the stage at which each patient was treated |
armleft |
The number of treatment left in the platform (>2) |
group |
The current stage |
variable.inf |
Fixeffect/Mixeffect for logistic model parameter |
reg.inf |
The information of how much accumulated information will be used |
A list of information require for the stan model including: zdropped: The vector of treatment index for each patient whose treatment arm is active at current stage. ydropped: The vector of outcome index for each patient whose treatment arm is active at current stage. Ndropped: The total number of patients that are treated with active treatment arms at current stage. group_indicator_dropped: The vector of stage index for each patient whose treatment arm is active at current stage. zlevel: The active treatment arm index at current stage xdummy: A design matrix transformed from zdropped and group_indicator_dropped for modelling
Ziyan Wang
stan.logisticmodeltrans(
z = c(1,2,1,2,2,1,2,1),
y = c(0,0,0,0,1,1,1,1),
randomprob = matrix(c( 0.5, 0.5), ncol = 2, dimnames = list(c("Stage1"), c("1", "2"))),
group_indicator = c(1,1,1,1,1,1,1,1),
armleft = 2,
group = 1,
variable.inf = "Fixeffect",
reg.inf = "main")