Skip to contents

Minimal fit class

MultistateModelFit.Rd

Minimal fit class

Minimal fit class

Public fields

model

The MultistateModel

data

A JointData object

info

Fit info.

Methods

Public methods

Method new()

Create model fit object

Usage

MultistateModelFit$new(data, stan_data, model, draws, info = NULL)

Arguments

data

Data used to create the fit.

stan_data

The used 'Stan' data list.

model

A MultistateModel

draws

A named list of rvars.

info

Fit info.

Method is_pk_only()

Is this a PK-only fit?

Usage

MultistateModelFit$is_pk_only()

Returns

logical

Method assert_hazard_fit()

Require that fit is not PK-only

Usage

MultistateModelFit$assert_hazard_fit()

Method mean_fit()

Create a reduced version with only a single draw, corresponding to the mean of original draws.

Usage

MultistateModelFit$mean_fit()

Returns

A new MultistateModelFit object.

Method is_point_estimate()

Check if fit is a point estimate

Usage

MultistateModelFit$is_point_estimate()

Returns

a logical value

Method get_data()

Extract Stan data list

Usage

MultistateModelFit$get_data()

Method draws_names()

Names of the draws list

Usage

MultistateModelFit$draws_names()

Method get_draws()

Extract draws as rvars

Usage

MultistateModelFit$get_draws(name = NULL)

Arguments

name

Param/quantity name

Method get_draws_of()

Draws in a raw array with same shape as Stan variable

Usage

MultistateModelFit$get_draws_of(name)

Arguments

name

Param/quantity name of x

Returns

Array with dimension c(ndraws(x), dim(x))

Method print()

Print the object

Usage

MultistateModelFit$print()

Returns

nothing

Method plot_basisfun()

Plot used basis functions (grid)

Usage

MultistateModelFit$plot_basisfun()

Method simulate_pk()

Simulate PK dynamics using fitted params.

Usage

MultistateModelFit$simulate_pk(
  oos = FALSE,
  data = NULL,
  L = 100,
  timescale = 24,
  n_prev = 3
)

Arguments

oos

Out-of-sample subjects?

data

Data for which to predict the concentration. If NULL, training data is used.

L

number of grid points for each subject

timescale

scale of time

n_prev

number of previous doses to show fit for

Method plot_pk()

Plot PK fit.

Usage

MultistateModelFit$plot_pk(
  max_num_subjects = 12,
  oos = FALSE,
  data = NULL,
  L = 100,
  timescale = 24,
  n_prev = 3,
  ci_alpha = 0.9
)

Arguments

max_num_subjects

Max number of subjects to show.

oos

Out-of-sample subjects?

data

Data for which to predict the concentration. If NULL, training data is used.

L

number of grid points for each subject

timescale

scale of time

n_prev

number of previous doses to show fit for

ci_alpha

Width of central credible interval.

Method plot_h0()

Plot baseline hazard distribution

Usage

MultistateModelFit$plot_h0(t = NULL, ci_alpha = 0.95)

Arguments

t

times where to evaluate the baseline hazards

ci_alpha

width of credible interval

Method h0_dist()

Baseline hazard distribution

Usage

MultistateModelFit$h0_dist(t = NULL, ci_alpha = 0.95)

Arguments

t

times where to evaluate the baseline hazards

ci_alpha

width of credible interval

Method covariate_effects()

Extract covariate effects

Currently not implemented for models that have a PK submodel.

Usage

MultistateModelFit$covariate_effects()

Returns

A data frame which has columns

  • covariate Name of the covariate

  • beta The covariate effect parameter estimate (rvar). NOTE: this is the regression coefficient for normalized covariates.

  • target_state Name of the target state. The corresponding beta is the covariate effect on all transitions that end in this target state.

Method log_z_pars()

Full names of parameters that start with log_z_.

Usage

MultistateModelFit$log_z_pars()

Method num_draws()

Get number of draws

Usage

MultistateModelFit$num_draws()

Method clone()

The objects of this class are cloneable with this method.

Usage

MultistateModelFit$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.