Package {impala}


Type: Package
Title: Bayesian Model Calibration
Version: 0.2.0
Description: Package provides tools for modular Bayesian model calibration. these tools allow for posterior exploration with sampling methods including tempering and adaptive Markov Chain Monte Carlo (MCMC). Allows for pooled calibration or hierarchal calibration of parameters. For more information see Francom et al., 2025 <doi:10.1137/24M1644092>.
License: MIT + file LICENSE
Encoding: UTF-8
Imports: cli, einsum, methods, parallel, progress, stats, utils
Suggests: fdasrvf, BASS, mvBayes, bayesplot, gridExtra, hexbin, knitr, rmarkdown, testthat (≥ 3.0.0)
Config/roxygen2/version: 8.1.1
Config/testthat/edition: 3
URL: https://github.com/sandialabs/rImpala
BugReports: https://github.com/sandialabs/rImpala/issues
VignetteBuilder: knitr
NeedsCompilation: no
Packaged: 2026-11-21 20:14:27 UTC; jdtuck
Author: J. Derek Tucker ORCID iD [aut, cre], Grant Hutchings ORCID iD [ctb], Gavin Collins [ctb], Sandia National Laboratories [cph, fnd]
Maintainer: J. Derek Tucker <jdtuck@sandia.gov>
Repository: CRAN
Date/Publication: 2026-11-21 20:40:02 UTC

Pooled Bayesian Model Calibration

Description

This function setups up calibration object

Usage

CalibSetup(bounds, constraint_func)

Arguments

bounds

a list with fields of variable names with values of two dimensional arrays where first element is lower bound and second element is upper bound.

constraint_func

function handle to constraint function on variables. Defaults to function on checking bounds

Value

An object of class 'CalibSetup' which is a list with the following components:

- 'nexp': number of experiments

- 'theta_fixed': named vector of parameters held fixed during calibration, 'NULL' until [fixTheta()] is called


PCA Based Model Emulator using BASS

Description

This function setups up emulator object

Usage

ModelBassPca_func(bmod, input_names, exp_ind = NULL, s2 = "MH")

Arguments

bmod

a object of the type 'bassBasis'

input_names

cell array of strings of input variable names

exp_ind

experiment indices (default: NULL)

s2

how to sample error variance (default: 'MH')

Value

An object of class 'ModelBassPca_func'


mvBayes Emulator for Functional Outputs (can use different BASS/BPPR type emulators)

Description

ModelmvBayes Handles larger-dimensional functional responses (e.g., on large spatial fields) using various inversion tricks. We require any other covariance e.g., from discrepancy, measurement error, and basis truncation error) to be diagonal. This function setups up emulator object.

Usage

ModelmvBayes(bmod, input_names, exp_ind = NULL, s2 = "MH", h = FALSE)

Arguments

bmod

a object of the type 'mvBayes'

input_names

cell array of strings of input variable names

exp_ind

experiment indices (default: NULL)

s2

how to sample error variance (default: 'MH')

h

h representation of warping function (default: FALSE)

Value

An object of class 'ModelmvBayes'


mvBayes Emulator for Functional Outputs (GP)

Description

ModelmvBayes_GP Handles larger-dimensional functional responses (e.g., on large spatial fields) using various inversion tricks. We require any other covariance e.g., from discrepancy, measurement error, and basis truncation error) to be diagonal. Built for a GP emulator.

Usage

ModelmvBayes_GP(bmod, input_names, exp_ind = NULL, s2 = "MH", h = FALSE)

Arguments

bmod

a object of the type 'mvBayes'

input_names

cell array of strings of input variable names

exp_ind

experiment indices (default: NULL)

s2

how to sample error variance (default: 'MH')

h

h representation of warping function (default: FALSE)

Details

This function setups up emulator object.

Value

An object of class 'ModelmvBayes_GP'


mvBayes elastic Emulator for Functional Outputs (can use different BASS/BPPR type emulators)

Description

ModelmvBayes_elastic Handles larger-dimensional functional responses (e.g., on large spatial fields) using various inversion tricks. We require any other covariance e.g., from discrepancy, measurement error, and basis truncation error) to be diagonal. Contains a change on the likelihood

Usage

ModelmvBayes_elastic(
  bmod,
  bmod_warp,
  input_names,
  exp_ind = NULL,
  s2 = "MH",
  h = FALSE
)

Arguments

bmod

a object of the type 'mvBayes' of aligned functions

bmod_warp

a object of the type 'mvBayes' of warping functions

input_names

cell array of strings of input variable names

exp_ind

experiment indices (default: NULL)

s2

how to sample error variance (default: 'MH')

h

h representation of warping function (default: FALSE)

Details

This function setups up emulator object.

Value

An object of class 'ModelmvBayes_elastic'


mvBayes elastic Emulator for Functional Outputs (can use different BASS/BPPR type emulators)

Description

mvBayes_elastic Handles larger-dimensional functional responses (e.g., on large spatial fields) using various inversion tricks. We require any other covariance e.g., from discrepancy, measurement error, and basis truncation error) to be diagonal. Contains a change on the likelihood and built for a GP emulator. This function setups up emulator object.

Usage

ModelmvBayes_elastic_GP(
  bmod,
  bmod_warp,
  input_names,
  exp_ind = NULL,
  s2 = "MH",
  h = FALSE
)

Arguments

bmod

a object of the type 'mvBayes' of aligned functions

bmod_warp

a object of the type 'mvBayes' of warping functions

input_names

cell array of strings of input variable names

exp_ind

experiment indices (default: NULL)

s2

how to sample error variance (default: 'MH')

h

h representation of warping function (default: FALSE)

Value

An object of class 'ModelmvBayes_elastic_GP'


Add joint theta prior

Description

This method adds a joint prior over multiple theta parameters

Usage

addJointThetaPrior(obj, pnames, log_density_fn)

Arguments

obj

'CalibSetup' Object

pnames

character vector of parameter names to apply joint prior to

log_density_fn

function that takes named list of parameter values and returns log-density

Details

'log_density_fn' receives a named list of parameter *vectors*, one element per temperature, and must return a vector of log densities of that same length. Write it vectorised, or wrap a scalar density in [mapply()].

Value

An object of class 'CalibSetup'

Examples

# Joint prior on t_1 and t_2 correlating them through a bivariate normal.
# params arrives as list(t_1 = <vector>, t_2 = <vector>), one entry per
# temperature, so the result must be a vector of the same length.
my_joint_prior <- function(params) {
  rho <- 0.5
  z <- (params$t_1^2 - 2 * rho * params$t_1 * params$t_2 + params$t_2^2) /
    (1 - rho^2)
  -0.5 * z - log(2 * pi * sqrt(1 - rho^2))
}

bounds = list()
bounds[['t_1']] = c(0, 1)
bounds[['t_2']] = c(0, 1)
setup <- CalibSetup(bounds, cf_bounds)
setup <- addJointThetaPrior(setup, c("t_1", "t_2"), my_joint_prior)

# returns one log density per temperature
my_joint_prior(list(t_1 = c(0.2, 0.5), t_2 = c(0.3, 0.6)))


Add theta prior

Description

This method adds a prior for the specified theta

Usage

addThetaPrior(
  obj,
  dist = "uniform",
  params = list(min = 0, max = 1),
  pname = NULL
)

Arguments

obj

'CalibSetup' Object

dist

string defining distribution e.g., 'normal', 'beta', etc.

params

list of parameters e.g., if dist='normal', params is a list with elements 'mean' and 'sd'

pname

name of parameter

Value

An object of class 'CalibSetup'


Add vector experiments

Description

This method adds vector experiments to calibration object

Usage

addVecExperiments(
  obj,
  yobs,
  model,
  sd_est,
  s2_df,
  s2_ind,
  sd_lower = NULL,
  sd_upper = NULL,
  meas_error_cor = NULL,
  theta_ind = NULL,
  D = NULL,
  discrep_tau = 1,
  wt = NULL
)

Arguments

obj

'CalibSetup' Object

yobs

a vector of the experiment or observation

model

emulator (currently expecting a object of class 'ModelBassPca_func' or 'ModelmvBayes')

sd_est

vector of initial standard deviation estimates, one per separately estimated measurement error variance. 'length(sd_est)' sets the number of variance groups.

s2_df

vector of inverse gamma prior degrees of freedom, one per variance group, so 'length(s2_df)' must equal 'length(sd_est)'. A value of zero selects a half-Cauchy prior instead; if any entry is zero the half-Cauchy is used for every group.

s2_ind

vector of 1-based variance group indices, one per element of 'yobs', mapping each observation to the variance it is drawn with. Must satisfy 'length(s2_ind) == length(yobs)' and 'max(s2_ind) <= length(sd_est)'.

sd_lower

optional vector of lower bounds on the learned standard deviations, one per variance group (default: 'NULL', meaning no lower bound)

sd_upper

optional vector of upper bounds on the learned standard deviations, one per variance group (default: 'NULL', meaning no upper bound)

meas_error_cor

measurement error correlation (default: 'NULL')

theta_ind

optional vector of 1-based calibration parameter indices, one per element of 'yobs', mapping each observation to the parameter vector \theta_j it is predicted with. Only used by [calibHier()] and [calibClust()], which give each index its own \theta_j; [calibPool()] uses a single \theta for everything. The indices must be '1, ..., k' with every value present. Default: 'NULL', meaning one \theta for the whole vector.

D

discrepancy basis (matrix of columns of basis, default: 'NaN')

discrep_tau

discrepancy sampling tau

wt

optional vector of non-negative observation weights, one per element of 'yobs' (default: 'NULL', meaning every weight is 1). Each observation's likelihood contribution is raised to the power of its weight, so a weight of 2 counts an observation twice and a weight of 0 drops it. Weights other than 1 need an emulator that implements them (currently [ModelBassPca_func()], or a custom model with 'supports_wt = TRUE').

Details

'sd_est', 's2_df', 'sd_lower' and 'sd_upper' are all indexed by variance group, while 's2_ind' is indexed by observation. A scalar 'sd_est' with 's2_ind = rep(1, length(yobs))' gives the usual single measurement error variance for the whole vector.

Any grouping of the observations is allowed. Passing 's2_ind = seq_along(yobs)' with 'sd_est' of that same length learns a separate standard deviation for every component of 'yobs'. Note that each variance is then informed by a single observation, so the prior set through 's2_df' and 'sd_est' does most of the work; use a larger 's2_df', or 'sd_lower'/'sd_upper', when the per component variances need to be constrained.

Value

An object of class 'CalibSetup'

Examples

set.seed(1)
ny <- 20
A <- matrix(stats::rnorm(ny * 2), ny, 2)
yobs <- as.numeric(A %*% c(0.45, 0.55)) + stats::rnorm(ny, 0, 0.05)

# a minimal linear emulator standing in for a fitted surrogate
mod <- structure(list(s2 = "gibbs", nd = 0, stochastic = FALSE,
                      discrep_cov = diag(ny) * 1e-12), class = "ExampleModel")

setup <- CalibSetup(list(t_1 = c(0, 1), t_2 = c(0, 1)), cf_bounds)

# two error groups: the first half of yobs shares one variance, the rest another
two_group <- addVecExperiments(setup, yobs, mod,
                               sd_est = c(0.05, 0.05),
                               s2_df  = c(2, 2),
                               s2_ind = rep(1:2, each = ny / 2))
two_group$ns2[[1]]

# one standard deviation per component of yobs, bounded away from zero
per_component <- addVecExperiments(setup, yobs, mod,
                                   sd_est   = rep(0.05, ny),
                                   s2_df    = rep(10, ny),
                                   s2_ind   = seq_along(yobs),
                                   sd_lower = rep(0.01, ny),
                                   sd_upper = rep(0.50, ny))
per_component$ns2[[1]]


Clustered Calibration

Description

Runs clustered Bayesian model calibration: the calibration parameter vectors \theta_j – one per value of 'theta_ind' in each [addVecExperiments()] call – are grouped by a Dirichlet process into at most 'nclustmax' clusters that share one parameter value. Cluster values are drawn from N(\theta_0, \Sigma_0) truncated to the constraints, with the hyperpriors of [setHierPriors()], and the concentration \eta has the Gamma prior of [setClusterPriors()]. Parameters live on the 0-1 scale of 'bounds'.

Usage

calibClust(setup)

Arguments

setup

an object of class 'CalibSetup', with [setHierPriors()] and [setClusterPriors()] called

Details

Memberships are updated one \theta_j at a time with the Chinese restaurant process weights, using the empty clusters (whose values are redrawn from N(\theta_0,\Sigma_0) every iteration) as the candidate new clusters; \eta is updated with the auxiliary variable scheme of Escobar and West (1995). Within a temperature the likelihood, the error variance prior and the N(\theta_0,\Sigma_0) density of the occupied clusters are tempered; the cluster process and the hyperpriors are not.

Discrepancy bases and [fixTheta()] are not supported; priors from [addThetaPrior()] are ignored.

Value

a list with the following elements - theta: 'nmcmc x ntemps x nclustmax x p' array of cluster values (0-1 scale); only clusters with members are meaningful - theta_hist: list over experiments of 'nmcmc x ntemps x ntheta x p' arrays holding the value each \theta_j took through its cluster - delta: list over experiments of 'nmcmc x ntemps x ntheta' arrays of cluster memberships (1-based) - eta: 'nmcmc x ntemps' matrix of the concentration parameter - theta0, Sigma0: hyperparameter samples, as in [calibHier()] - s2: list over experiments of error variance samples - count: accepted cluster value updates per temperature - count_s2: acceptance counts for Metropolis error variance updates - count_temper: 'ntemps x ntemps' accepted tempering swaps - pred_curr: final predictions for each experiment - llik: log likelihood of the cold chain at each iteration - nclustmax: the maximum number of clusters - theta_native: list over experiments of lists over j of data frames of cold chain \theta_j samples on the native scale - theta0_native, theta_parent_native: as in [calibHier()]


Hierarchical Calibration

Description

Runs hierarchical Bayesian model calibration with adaptive MCMC, tempering and a joint decorrelation step. Every calibration parameter vector \theta_j – one per value of 'theta_ind' in each [addVecExperiments()] call – gets its own value, and the \theta_j are tied together by \theta_j \sim N(\theta_0, \Sigma_0), truncated to the constraints, with the hyperpriors set by [setHierPriors()]. Parameters live on the 0-1 scale of 'bounds'.

Usage

calibHier(setup)

Arguments

setup

an object of class 'CalibSetup', with [setHierPriors()] called

Details

Within a temperature the likelihood, the N(\theta_0,\Sigma_0) density of each \theta_j and the error variance prior are tempered, while the hyperpriors on \theta_0 and \Sigma_0 are not; the Gibbs updates, the decorrelation step and the swaps all target that same tempered posterior. Discrepancy bases are not supported here.

Error variance groups ('s2_ind') may be shared across several \theta_j. Parameters fixed with [fixTheta()] are not supported, and priors from [addThetaPrior()] are ignored in favour of the hierarchical prior.

Value

a list with the following elements - theta: list over experiments of 'nmcmc x ntemps x ntheta x p' arrays of \theta_j samples (0-1 scale) - theta0: 'nmcmc x ntemps x p' array of \theta_0 samples - Sigma0: 'nmcmc x ntemps x p x p' array of \Sigma_0 samples - s2: list over experiments of error variance samples - count: list over experiments of 'ntemps x ntheta' acceptance counts - count_s2: acceptance counts for Metropolis error variance updates - count_decor: 'ntemps x p' acceptance counts of the decorrelation step - count_temper: 'ntemps x ntemps' accepted tempering swaps - cov_theta_cand: adaptive Metropolis state for theta - cov_ls2_cand: adaptive Metropolis state for the log error variances - pred_curr: final predictions for each experiment - llik: log likelihood of the cold chain at each iteration - theta_native: list over experiments of lists over j of data frames of cold chain \theta_j samples on the native scale - theta0_native: data frame of cold chain \theta_0 samples on the native scale - theta_parent_native: data frame of draws from N(\theta_0, \Sigma_0) (truncated to the constraints) at each cold chain sample, i.e. the parameter of a new, unobserved experiment


Pool Calibration

Description

This function runs pooled Bayesian Model Calibration with adaptive MCMC, tempering, and decorrelation steps. input theta will be normalized to 0-1 and sampled from uniform priors

Usage

calibPool(setup)

Arguments

setup

an object of class 'CalibSetup'

Details

Parameters fixed with [fixTheta()] are held at their given values throughout: they keep their column in 'theta' and 'theta_native', constant at the fixed value, while the Metropolis proposal and the decorrelation step act on the free parameters only.

Value

a list with the following elements - theta: mcmc samples of variables - s2: mcmc samples of error variance - count: number of counts of acceptance - count_s2: number of counts of acceptance on error variance - count_decor: number of times decorrelation occurred - cov_theta_cand: final theta covariance - cov_ls2_cand: final error covariance - pred_curr: current emulator predictions - discrep_vars: discrepancy coefficients - llik: log likelihood - theta_native: mcmc samples of variables in native scale - theta_fixed: named vector of parameters held fixed, or 'NULL' if none


Compare to bounds

Description

This function compares variables to bounds

Usage

cf_bounds(x, bounds)

Arguments

x

list of of parameters

bounds

list of bounds for each parameter that is a two parameter vector with high and low

Value

a vector of 'TRUE' or 'FALSE' if the values are within the bounds


Sample discrepancy coefficients

Description

Generic that draws discrepancy basis coefficients for an emulator model with a discrepancy basis attached.

Usage

discrep_sample(obj, ...)

Arguments

obj

an emulator model object

...

additional arguments passed to the method, typically the observations, the prediction, the marginal covariance, and the inverse temperature

Value

a vector of sampled discrepancy coefficients


evalm constructor

Description

Default constructor for evalm class

Usage

evalm(obj, ...)

Arguments

obj

evalm object

...

additional arguments passed to method

Value

An object of class 'evalm'


Fix calibration parameters

Description

This method holds one or more calibration parameters at fixed values, so that [calibPool()] samples the conditional posterior of the remaining parameters. This supports conditional inference and cut-Bayes style analyses, where some inputs are treated as known, or as determined by a separate module, rather than learned from the data at hand.

Usage

fixTheta(obj, pname, value)

Arguments

obj

'CalibSetup' Object

pname

character vector of parameter names to fix; each must appear in 'names(obj$bounds)'

value

numeric vector of values to fix the parameters at, on the native scale of 'obj$bounds' and the same length as 'pname'

Details

Fixed parameters keep their column in the 'theta' and 'theta_native' output of [calibPool()]; that column is simply constant at the fixed value. The adaptive Metropolis proposal and the decorrelation step both act on the free parameters only, so fixing a parameter does not degrade the mixing of the others.

Calling 'fixTheta()' again for a parameter that is already fixed replaces its value. Use [unfixTheta()] to return a parameter to being sampled.

A prior added with [addThetaPrior()] or [addJointThetaPrior()] for a parameter that is fixed contributes a constant to the log posterior and so has no effect on the samples; a warning is issued when that is detected.

Fixing every parameter is allowed: [calibPool()] then samples the error variances and any discrepancy terms with 'theta' held at the fixed values.

Value

An object of class 'CalibSetup'

Examples

bounds = list()
bounds[['t_1']] = c(0, 1)
bounds[['t_2']] = c(-2, 2)
setup <- CalibSetup(bounds, cf_bounds)

# hold t_2 at 0.5 and calibrate t_1 conditional on it
setup <- fixTheta(setup, "t_2", 0.5)
setup$theta_fixed

# several at once
setup <- fixTheta(setup, c("t_1", "t_2"), c(0.25, -1))
setup$theta_fixed

# and back to sampling both
setup <- unfixTheta(setup)
setup$theta_fixed


Generate a proposal

Description

Generic that draws a Metropolis candidate from an adaptive proposal covariance.

Usage

gen_cand(obj, ...)

Arguments

obj

an adaptive covariance object, e.g. from 'AMcov_pool'

...

additional arguments passed to the method, typically the chain of samples and the current iteration

Value

a matrix of candidate values, one row per temperature


Likelihood covariance inverse

Description

Generic that builds the inverse and log determinant of the marginal likelihood covariance for an emulator model.

Usage

lik_cov_inv(obj, ...)

Arguments

obj

an emulator model object

...

additional arguments passed to the method, typically the vector of error variances

Value

a list with elements 'inv' (the inverse covariance) and 'ldet' (its log determinant)


Log likelihood

Description

Generic for the log likelihood of an emulator model. Methods are provided for each supported emulator class; supply your own to calibrate with a custom emulator.

Usage

llik(obj, ...)

Arguments

obj

an emulator model object

...

additional arguments passed to the method, typically the observations, the emulator prediction, and the marginal covariance returned by [lik_cov_inv()]

Value

the log likelihood, a scalar


Set Cluster Priors

Description

Sets the Dirichlet process settings used by [calibClust()]. The \theta_j are grouped into at most 'nclustmax' clusters that share a parameter value. The concentration parameter \eta has a Gamma('eta_prior_shape', 'eta_prior_rate') prior; larger \eta favours more clusters. Call [setHierPriors()] as well, since the cluster values are drawn from N(\theta_0, \Sigma_0).

Usage

setClusterPriors(
  obj,
  nclustmax = NULL,
  eta_prior_shape = 2,
  eta_prior_rate = 0.1,
  scale_proposal = FALSE
)

Arguments

obj

'CalibSetup' object

nclustmax

maximum number of clusters (default: 'NULL', meaning 'max(sum(ntheta), 10)', where 'ntheta' counts the parameter vectors set up through 'theta_ind')

eta_prior_shape

shape of the Gamma prior on \eta (default: 2)

eta_prior_rate

rate of the Gamma prior on \eta (default: 0.1)

scale_proposal

if ‘TRUE', divide each cluster’s proposal covariance by its number of members, which can help when cluster sizes are very uneven (default: 'FALSE')

Value

An object of class 'CalibSetup'


Set Hierarchical Priors

Description

Sets the hyperpriors used by [calibHier()] and [calibClust()]. Each parameter vector \theta_j (one per value of 'theta_ind', see [addVecExperiments()]) is drawn from N(\theta_0, \Sigma_0), with \theta_0 \sim N('theta0_prior_mean', 'theta0_prior_cov') and \Sigma_0 \sim Inverse-Wishart('Sigma0_prior_df', 'Sigma0_prior_scale'). All of these live on the 0-1 scale the sampler works on, not the native scale of 'bounds'.

Usage

setHierPriors(
  obj,
  theta0_prior_mean = NULL,
  theta0_prior_cov = NULL,
  Sigma0_prior_df = NULL,
  Sigma0_prior_scale = NULL,
  theta0_start = NULL
)

Arguments

obj

'CalibSetup' object

theta0_prior_mean

prior mean of \theta_0, length 'p' (default: 'rep(0.5, p)')

theta0_prior_cov

prior covariance of \theta_0, 'p x p' (default: 'diag(p) * 100', effectively flat on the unit cube)

Sigma0_prior_df

Inverse-Wishart degrees of freedom for \Sigma_0; must exceed 'p + 1' so the prior mean exists. Larger values pull the \theta_j together (default: 'p + 20')

Sigma0_prior_scale

Inverse-Wishart scale matrix, 'p x p'. The prior mean of \Sigma_0 is 'Sigma0_prior_scale / (Sigma0_prior_df - p - 1)' (default: 'diag(p) * 0.1^2')

theta0_start

optional 'ntemps x p' matrix of 0-1 scale starting values for \theta_0, one row per temperature (default: 'NULL', a random start satisfying the constraints)

Value

An object of class 'CalibSetup'


Set MCMC Parameters

Description

This function setups up MCMC parameters for adaptive MCMC, also includes tempering and decorrelation steps

Usage

setMCMC(
  obj,
  nmcmc,
  nburn = 0,
  thin = 1,
  decor = 100,
  start_var_theta = 1e-08,
  start_tau_theta = 0,
  start_var_ls2 = 1e-05,
  start_tau_ls2 = 0,
  start_adapt_iter = 300
)

Arguments

obj

'calibPool' object

nmcmc

number of mcmc iterations

nburn

number of mcmc burn in iterations (default: 0)

thin

number of samples to thin (default: 1)

decor

number of mcmc iterations before decorrelation step (default: 100)

start_var_theta

start variance of theta proposal (default: 1e-8)

start_tau_theta

start tau of theta (default: 0)

start_var_ls2

start variance of sigma proposal (default: 1e-5)

start_tau_ls2

start tau of sigma (default: 0)

start_adapt_iter

number of iterations before adaption (default: 300)

Value

An object of class 'CalibSetup'


Set Temperature Ladder

Description

This function setups up temperature ladder for tempering

Usage

setTemperatureLadder(obj, temperature_ladder, start_temper = 1000)

Arguments

obj

CalibSetup Object

temperature_ladder

array of temperatures

start_temper

what MCMC sample to start tempering (default: 1000)

Value

An object of class 'CalibSetup'


Advance a stochastic emulator

Description

Generic that advances an emulator one MCMC step, used for emulators that carry their own posterior samples.

Usage

step_m(obj, ...)

Arguments

obj

an emulator model object

...

additional arguments passed to the method

Value

the updated emulator model object


Transform parameters

Description

Transforms variables from the scale 0 to 1 to variable provided by bounds

Usage

tran_unif(th, bounds, names)

Arguments

th

list of of parameters

bounds

list of bounds for each parameter that is a two parameter vector with high and low

names

vector of variable names

Value

a list of parameters


Unfix calibration parameters

Description

This method releases parameters fixed by [fixTheta()] so that [calibPool()] samples them again.

Usage

unfixTheta(obj, pname = NULL)

Arguments

obj

'CalibSetup' Object

pname

character vector of parameter names to release, or 'NULL' (the default) to release every fixed parameter

Value

An object of class 'CalibSetup'


Update adaptive covariance

Description

Generic that updates the running mean and covariance of an adaptive Metropolis proposal.

Usage

update_m(obj, ...)

Arguments

obj

an adaptive covariance object, e.g. from 'AMcov_pool'

...

additional arguments passed to the method, typically the chain of samples and the current iteration

Value

the updated adaptive covariance object


Update proposal scaling

Description

Generic that applies diminishing adaptation to the proposal scale based on the recent acceptance rate.

Usage

update_tau(obj, ...)

Arguments

obj

an adaptive covariance object, e.g. from 'AMcov_pool'

...

additional arguments passed to the method, typically the current iteration

Value

the updated adaptive covariance object