Analysis / inference / implementation capabilities
Source:R/inference.R
interactmeta_capability_matrix.RdThree distinct axes:
analysis — DA, AD, SWADA, or CAMS
inference — Bayesian or frequentist
engine — computational backend that implements that inference
CAMS is one statistical model. Bayesian CAMS is implemented using
bayesmeta (DIRECT); frequentist CAMS is implemented using metafor
(ML/REML). DA and AD are not inherently frequentist. Bayesian inference
does not imply CAMS information-fraction functionality. IF quantities and
IF plots are CAMS-only.
Value
A data frame with one row per public analysis × engine × method combination, including heterogeneity, posterior, IF functionality, and IF plots. Combinations that are theoretically possible but not part of the public API are omitted.
Examples
interactmeta_capability_matrix()
#> analysis engine method algorithm inference heterogeneity posterior
#> 1 cam metafor REML optimization frequentist TRUE FALSE
#> 2 cam metafor ML optimization frequentist TRUE FALSE
#> 3 cam bayesmeta bayes quadrature bayesian TRUE TRUE
#> 4 da metafor REML optimization frequentist TRUE FALSE
#> 5 da metafor ML optimization frequentist TRUE FALSE
#> 6 da bayesmeta bayes quadrature bayesian TRUE TRUE
#> 7 ad metafor REML optimization frequentist TRUE FALSE
#> 8 ad metafor ML optimization frequentist TRUE FALSE
#> 9 ad bayesmeta bayes quadrature bayesian TRUE TRUE
#> 10 swada metafor REML optimization frequentist TRUE FALSE
#> 11 swada metafor ML optimization frequentist TRUE FALSE
#> ifrac ifrac_plots interaction subgroups
#> 1 TRUE TRUE TRUE TRUE
#> 2 TRUE TRUE TRUE TRUE
#> 3 TRUE TRUE TRUE TRUE
#> 4 FALSE FALSE TRUE TRUE
#> 5 FALSE FALSE TRUE TRUE
#> 6 FALSE FALSE TRUE TRUE
#> 7 FALSE FALSE TRUE FALSE
#> 8 FALSE FALSE TRUE FALSE
#> 9 FALSE FALSE TRUE FALSE
#> 10 FALSE FALSE TRUE TRUE
#> 11 FALSE FALSE TRUE TRUE