formula
|
Either an R formula in the form response/DV ~ var1 + var2 etc. or formula object as created/called by the brms brms::bf function.
|
data
|
An R data frame or tibble containing the variables in the formula
|
true_bounds
|
If the true bounds of the outcome/response don’t exist in the data, pass a length 2 numeric vector of the minimum and maximum bounds to properly normalize the outcome/response
|
phi_reg
|
Whether you are including a linear model predicting the dispersion parameter, phi, and/or for the response. If you are including models for both, pass option ‘both’. If you only have an intercept for the outcome (i.e. a 1 in place of covariates), pass ‘only’. If you only have intercepts for phi (such as a varying intercepts/random effects) model, pass the value "intercepts". To set priors on these intercepts, use the extra-prior option with the brms::set_prior function (class="sd"). If no model of any kind for phi, the default, pass ‘none’.
|
use_brm_multiple
|
(T/F) Whether the model should use brms::brm_multiple for multiple imputation over multiple dataframes passed as a list to the data argument
|
coef_prior_mean
|
The mean of the Normal distribution prior on the regression coefficients (for predicting the mean of the response). Default is 0.
|
coef_prior_SD
|
The SD of the Normal distribution prior on the regression coefficients (for predicting the mean of the response). Default is 5, which makes the prior weakly informative on the logit scale.
|
intercept_prior_mean
|
The mean of the Normal distribution prior for the intercept. By default is NULL, which means the intercept receives the same prior as coef_prior_mean. To zero out the intercept, set this parameter to 0 and coef_prior_SD to a very small number (0.01 or smaller). NOTE: the default intercept in brms is centered (mean-subtracted) by default. To use a traditional intercept, either add 0 + Intercept to the formula or specify center=FALSE in the bf formula function for brms. See brms::brmsformula() for more info.
|
intercept_prior_SD
|
The SD of the Normal distribution prior for the intercept. By default is NULL, which means the intercept receives the same prior SD as coef_prior_SD.
|
phi_prior
|
The mean parameter of the exponential prior on phi, which determines the dispersion of the beta distribution. The default is .1, which equals a mean of 10 and is thus weakly informative on the interval (0.4, 30). If the response has very low variance (i.e. tightly) clusters around a specific value, then decreasing this prior (and increasing the expected value) may be helpful. Checking the value of phi in the output of the model command will reveal if a value of 0.1 (mean of 10) is too small.
|
dirichlet_prior
|
A vector of three integers corresponding to the prior parameters for the dirchlet distribution (alpha parameter) governing the location of the cutpoints between the components of the response (continuous vs. degenerate). The default is 1 which puts equal probability on degenerate versus continuous responses. Likely only needs to be changed in a repeated sampling situation to stabilize the cutpoint locations across samples.
|
phi_coef_prior_mean
|
The mean of the Normal distribution prior on the regression coefficients for predicting phi, the dispersion parameter. Only useful if a linear model is being fit to phi. Default is 0.
|
phi_coef_prior_SD
|
The SD of the Normal distribution prior on the regression coefficients for predicting phi, the dispersion parameter. Only useful if a linear model is being fit to phi. Default is 5, which makes the prior weakly informative on the exponential scale.
|
phi_intercept_prior_mean
|
The mean of the Normal distribution prior for the phi (dispersion) regression intercept. By default is NULL, which means the intercept receives the same prior as phi_coef_prior_mean. To zero out the intercept, set this parameter to 0 and phi_coef_prior_SD to a very small number (0.01 or smaller).
|
phi_intercept_prior_SD
|
The SD of the Normal distribution prior for the phi (dispersion) regression intercept. By default is NULL, which means the intercept receives the same prior SD as phi_coef_prior_SD.
|
|
|
An additional prior, such as a prior for a specific regression coefficient, added to the outcome regression by passing one of the brms functions brms::set_prior or brms::prior_string with appropriate values.
|
manual_prior
|
If you want to set your own custom priors with brms, use this option to pass any valid brms priors such as those created with brms::set_prior or brms::prior_string. Note that this option replaces any other priors set. Useful especially when doing something unorthodox like modeling cutpoints.
|
init
|
This parameter is used to determine starting values for the Stan sampler to begin Markov Chain Monte Carlo sampling. It is set by default at 0 because the non-linear nature of beta regression means that it is possible to begin with extreme values depending on the scale of the covariates. Setting this to 0 helps the sampler find starting values. It does, on the other hand, limit the ability to detect convergence issues with Rhat statistics. If that is a concern, such as with an experimental feature of brms, set this to “random” to get more robust starting values (just be sure to scale the covariates so they are not too large in absolute size).
|
return_stancode
|
If TRUE, will pass back the only the Stan code for the model as a character vector rather than fitting the model.
|
…
|
All other arguments passed on to the brm function
|