View source: R/newMLEfunction.R
Generate a list of parameters controllong the stochastic gradient process of the roiMLE
function
1 2 3 4 | roi_mle_control(grad_iterations = 2100, step_size_coef = 0.5,
step_rate = 0.55, samp_per_iter = 20, grad_delay = NULL,
assume_convergence = NULL, impute_boundary = c("smooth", "neighbors",
"none", "mean"), mahal_weight = 0.2, RB_mult = 1)
|
grad_iterations |
the number of stochastic gradient steps to take |
step_size_coef |
step size coefficients for stochastic gradient step. Best left unchanged. |
step_rate |
the rate at which the stochastic gradient steps size should decrease as a function of the number of steps already taken. |
samp_per_iter |
the number of slice MH samples to take for computing the stochastic gradient estimate in each stochastic gradient step |
grad_delay |
the number of iterations to wait before starting to decrease the stochastic gradient step size |
assume_convergence |
after how many gradient steps should we assume
convergence? The final MLE estimate will be the average of the last
|
impute_boundary |
the boundary imputation method to use. See description for details |
RB_mult |
adjusts the Robins-Monroe step sizes when computing profile-likelihood confidence intervals. |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.