indirect {indirect} | R Documentation |
indirect: A package for assisting indirect elicitation of priors for generalised linear models.
Description
The indirect
package provides three categories of functions: elicitation
functions, fitting functions and visualisation functions.
Elicitation functions
These are the functions that are used to
record expert opinion. This is where edits will be made and so on. The key
function is designLink
, which defines a list object that contains
information about the design and elicitation. The elicitations are recorded and updated
via function elicitPt
.
Fitting functions
These are generally helper functions except for the function
muSigma
that is used for estimating the mean vector and covariance
matrix of the unknown coefficients for the multivariate normal prior. Helper functions
include mV
for the elicited moments of conditional means priors.
Visualisation functions
These are functions for visualisation. The
core function is plotDesignPoint
.
References
Hosack, G. R., Hayes, K. R., & Barry, S. C. (2017). Prior elicitation for Bayesian generalised linear models with application to risk control option assessment. Reliability Engineering and System Safety, 167:351-361. doi:10.1016/j.ress.2017.06.011