prior_sample {survextrap} | R Documentation |
Sample from the joint prior of parameters in a survextrap model
Description
Draws a sample from the joint prior distribution of the parameters
governing a survextrap model for given covariates. This is used,
for example, in prior_sample_hazard
, to visualise the
prior distribution around hazard curves implied by a particular
M-spline model and parameter priors.
Usage
prior_sample(
mspline,
coefs_mean = NULL,
prior_hsd = p_gamma(2, 1),
prior_hscale,
smooth_model = "exchangeable",
prior_loghr = NULL,
formula = NULL,
cure = NULL,
nonprop = NULL,
newdata = NULL,
newdata0 = NULL,
prior_hrsd = NULL,
prior_cure = NULL,
prior_logor_cure = NULL,
nsim = 100
)
Arguments
mspline |
A list of control parameters defining the spline model.
If there are external data, and both
|
coefs_mean |
Spline basis coefficients that define the prior
mean for the hazard function. By default, these are set to values
that define a constant hazard function (see
|
prior_hsd |
Gamma prior for the standard deviation that
controls the variability over time (or smoothness) of the hazard
function. This should be a call to |
prior_hscale |
Prior for the baseline log hazard scale
parameter ( Note that "Baseline" is defined by the continuous covariates taking a value of zero and factor covariates taking their reference level. To use a different baseline, the data should be transformed appropriately beforehand, so that a value of zero has a different meaning. For continuous covariates, it helps for both computation and interpretation to define the value of zero to denote a typical value in the data, e.g. the mean. |
smooth_model |
The default The alternative In non-proportional hazards models, setting |
prior_loghr |
Priors for log hazard ratios. This should be a
call to The default is |
formula |
A model formula with no response, defining the covariates on the hazard scale. |
cure |
A model formula with no response, giving any covariates on the cure proportion. |
nonprop |
A model formula with no response, defining any covariates affecting the spline basis coefficients, which gives a nonproportional hazards model. |
newdata |
A data frame with one row, containing variables in the model formulae. Samples will then be drawn, for any covariate-dependent parameters, with covariates set to the values given here. |
newdata0 |
A data frame with one row, containing "reference"
values of variables in the model formulae. The hazard ratio
between the hazards at |
prior_hrsd |
Prior for the standard deviation parameters that
smooth the non-proportionality effects over time in
non-proportional hazards models. This should be a call to
|
prior_cure |
Prior for the baseline cure probability. This should be a
call to |
prior_logor_cure |
Priors for log odds ratios on cure probabilities.
This should be a call to |
nsim |
Number of simulations to draw |
Value
prior_sample_hazard
returns a data frame of the
samples, and plot_prior_hazard
generates a plot. No
customisation options are provided for the plot function, which
is just intended as a quick check.
A list with components:
alpha
: Baseline log hazard scale parameter (log(eta)
in the notation of the manual). For models with covariates, this is at the covariate values supplied in X
, or at zero if X
is not supplied.
hscale
: Baseline hazard scale parameter (eta
).
coefs
: Spline coefficients. For non-proportional hazards model with covariates, these are returned at the suppled value of X
, or at values of zero if X
is not supplied.
beta
: Multinomial logit-transformed spline coefficients.
hsd
: Smoothing standard deviation for spline coefficients.
If X0
is supplied, then alpha0
, hscale0
, beta0
, coefs0
are also returned, representing reference covariate values.
pcure
is returned in cure models (the cure probability).