simulationABC {poolABC}

Perform an Approximate Bayesian Computation simulation study

Description

Perform a leave-one-out cross validation for ABC via subsequent calls to the singleABC() function.

Usage

simulationABC(
  params,
  sumstats,
  limits,
  nval,
  tol,
  method,
  parallel = FALSE,
  ncores = NA
)

Arguments

params

is a vector or matrix of simulated parameter values i.e. numbers from the simulations. Each row or vector entry should be a different simulation and each column of a matrix should be a different parameter. This is the dependent variable for the regression, if a regression step is performed.

sumstats

is a vector or matrix of simulated summary statistics. Each row or vector entry should be a different simulation and each column of a matrix should be a different statistic. These act as the independent variables if a regression step is performed.

limits

is a matrix with two columns and as many rows as there are parameters. Each row should contain the minimum value of the prior for a given parameter in the first column and the maximum value in the second column.

nval

size of the cross-validation sample i.e. how many different evaluations should be performed. Each evaluation corresponds to a different target for the parameter estimation.

tol

is the tolerance rate, indicating the required proportion of points accepted nearest the target values.

method

either "rejection" or "regression" indicating whether a regression step should be performed during ABC parameter estimation.

parallel

logical, indicating whether this function should be run using parallel execution. The default setting is FALSE, meaning that this function will utilize a single core.

ncores

a non-negative integer that is required when parallel is TRUE. It specifies the number of cores to use for parallel execution.

Details

This function allows users to evaluate the impact of different tolerance rate on the quality of the estimation with ABC and whether a local linear regression algorithm improves the estimates. In subsequent steps, different point estimates of the posterior estimates can be compared with the true values, allowing the users to select the point estimate that leads to lower errors. Thus, performing a leave-one-out cross validation aids in selecting which point estimate is best - the mean, median or mode.

Value

a list with the following elements:

true

The parameter values of the simulations that served as validation.

rej

a list with the estimated parameter values under the rejection algorithm and using three different point estimates: mode, median and mean. The final entry of the list is the prediction error for each parameter, considering each of those point estimates as the estimated value.

reg

if method is "regression" then this is a list with the estimated parameter values under the regression algorithm and using three different point estimates: mode, median and mean. The final entry of the list is the prediction error for each parameter, considering each of those point estimates as the estimated value.

Examples

# load the matrix with parameter values
data(params)
# load the matrix with simulated parameter values
data(sumstats)
# load the matrix with the prior limits
data(limits)

# perform a leave-one-out cross validation for ABC
simulationABC(params = params, sumstats = sumstats, limits, nval = 10,
tol = 0.01, method = "regression")