R Package
An R package, SHELF, is available on CRAN, and a developement version is on GitHub. The package includes both Shiny apps and command line implementation for all the methods described in the SHELF documentation.
Shiny apps
These apps are available online, but are also all included in the SHELF R package. Online access to these apps is time-limited (they are hosted on RStudio’s shinyapps.io service). If you want to use the apps in an actual elicitation workshop, we strongly encourage you to use the offline versions in the SHELF package.
Single expert
Eliciting a distribution from a single expert
Multiple experts
Eliciting individual distributions from multiple experts
Bivariate elicitation
Eliciting a bivariate distribution using a Gaussian copula
Dirichlet elicitation
Eliciting a Dirichlet distribution for a set of proportions constrained to sum to 1.
Extension method - continuous
Eliciting a distribution for a continuous target variable $X$ via a distribution for a continuous extension variable $Y$ and a conditional distribution for $X|Y$.
Extension method - discrete
Eliciting a distribution for a continuous target variable $X$ via a distribution for a discrete extension variable $Y$ and a conditional distribution for $X|Y$. Can be used to elicit multimodal distributions.
Other elicitation methods
We have been involved in developing other elicitation methods, outside of the SHELF project. There is a paper and supporting app/code for each method, but templates for conducting the elicitation are not available within SHELF
-
Elicitation for survival and time-to-event outcomes
- Ren, S. and Oakley, J. E. (2014). Assurance calculations for planning clinical trials with time-to-event outcomes. Statistics in Medicine 33(1), 31-45.
- Download supporting R code
-
Eliciting proper priors for heterogeneity in random-effects meta analysis
- Ren, S., Oakley, J. E. and Stevens, J. W. (2018). Incorporating genuine prior information about between-study heterogeneity in random effects pairwise and network meta-analyses. Medical Decision Making, 38 (4), 531-542.
- Please see the appendix for details of supporting software.
-
Eliciting distributions for population variances
- Alhussain, Z. A. and Oakley, J. E. (2020). Assurance for clinical trial design with normally distributed outcomes: eliciting uncertainty about variances. Pharmaceutical Statistics, 19, 827-839.
- Supporting shiny app
- R package (assurance) on GitHub
MATCH elicitation tool
As part of the MATCH project, David Morris, Jeremy Oakley and John Crowe produced a web-based elictation tool which is based on an earlier version of the SHELF R code. A user guide was produced as part of the PCOD+ project, sponsored by the Office for Naval Research.
MATCH uncertainty elicitation tool
Eliciting a distribution from a single expert
MATCH user guide
MATCH user guide (pdf)