Learn R Programming

BCEA (version 2.1-0)

evppi: Expected Value of Perfect Partial Information (EVPPI) for selected parameters

Description

Computes the Expected Value of Perfect Partial Information (EVPPI) with respect to a given parameter.

Usage

evppi(parameters,inputs,he,n.blocks=NULL,n.seps=NULL,method=c("so","sal"))

## S3 method for class 'default': evppi(parameters,inputs,he,n.blocks=NULL,n.seps=NULL,method=c("so","sal"))

Arguments

parameters
A string (or vector of strings) containing the names of the parameters for which the EVPPI should be computed. If more than one parameter is specified, the EVPPI will be computed for each separately.
inputs
A matrix containing the simulations for all the parameters monitored by the call to JAGS or BUGS. The matrix should have column names matching the names of the parameters and the values in the vector parameter should match at least one of those values.
he
A bcea object (the result of the call to the function bcea).
n.blocks
The number of blocks in which the matrix of inputs should be decomposed to perform the calculations. The number of blocks must be such that the ratio of the number of MCMC simulations to the number of blocks is an integer number. The choice of this para
n.seps
The number of points defining the separation in the matrix of inputs. The available number of separation points is 1 (default) or 2. This parameter is only used if method = "sal".
method
The method used to perform the EVPPI calculation. Options are Strong and Oakley ("so") or Sadatsafavi et al ("sal").

Value

  • evppiThe computed values of evppi for all values of the parameter of willingness to pay
  • parametersThe string vector with the names of the parameters for which the EVPPI is computed
  • kThe vector of values for the willingness to pay
  • eviThe vector of values for the overall EVPI
  • methodThe string reporting the method selected for calculations

Details

Computes the Expected Value of (Partial) Partial Information with respect to a given parameter, using the algorithm for fast computation of Strong and Oakley (2013) or of Sadatsafavi et al. (2013).

References

Strong, M. and Oakley, J.E. (2013). An efficient method for computing partial expected value of perfect information for correlated inputs

Sadatsafavi et al. (2013). Need for Speed: An efficient algorithm for calculation of single-parameter expected value of partial perfect information

Baio G. (2012). Bayesian Methods in Health Economics. CRC/Chapman Hall, London

See Also

plot.evppi, bcea