The Jump Over move is accompanied with proposing new slopes that ensure a continuous ER.
1 2 3 | JumpOverCoef(current_coefs, prop_cuts, cuts, curr_exper_same,
prop_exper_same, curr_exper_comb, prop_exper_comb, curr_exper_split,
prop_exper_split, jump_slope_tune)
|
current_coefs |
The current coefficients of the MCMC. Three dimensional array with dimensions corresponding to exposure/outcome model, experiment, and coefficients (intercept, slope, covariates). |
prop_cuts |
Numeric vector with the K locations of the proposed experiment configuration. |
cuts |
Numeric vector with the K locations of the current experiment configuration. |
curr_exper_same |
Numeric vector. Indices of experiments in the current experiment configuration that did not get combined or split. They should be in the same order as prop_exper_same. |
prop_exper_same |
Numeric vector with the indices of the experiments in the proposed experiment configuration that did not get combined or split. |
curr_exper_comb |
Numeric vector of length two with entries the indices of the experiments with respect to the current experiment configuration that were proposed to be combined. |
prop_exper_comb |
Numeric. Index of the experiment in the proposed experiment configuration corresponding to the combined experiment. |
curr_exper_split |
Numeric. Index of the experiment in the current experiment configuration that is proposed to be split. |
prop_exper_split |
Numeric vector of length two. Indices of the experiments in the proposed experiment configuration that are the result of a split in an experiment. |
jump_slope_tune |
Numeric. The standard deviation of the proposal on slopes for the two experiments in the proposed experiment configuration that are the result of an experiment's split. |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.