Description Details Author(s) References Examples
The main function sensitivity_learning_table
() provides posterior, sensitivity and learning estimates
for the Bayesian normal-normal hierarchical model used for Bayesian meta-analysis
under four different heterogeneity priors (half-normal, half-Cauchy, exponential, Lomax).
The more advanced function sensitivity_learning_table_flexible
()
enables a flexible choice of several parameters
and supports computation of the reference within-study standard deviation
based on both the geometric mean and a weighted harmonic mean.
In order to unify notation, the heterogeneity priors are defined
as scaled distributions tau ~ A_0 |X|,
where A_0 is a scale parameter and
X is the standard form of the distribution.
The methodology implemented is proposed in Ott et al. (2020).
The function pri_par_adjust_dynamic
() implements the novel heterogeneity prior adjustment
with respect to the relative latent model complexity (RLMC).
The function pri_par_adjust_static
() implements the standard approach to
heterogeneity prior tail adjustment (Spiegelhalter et al. 2004).
Package: sl4bayesmeta
Type: Package
Title: Sensitivity and learning for Bayesian meta-analysis
Version: 0.3-1
Date: 2020-02-18
Author: Manuela Ott [aut, cre], Malgorzata Roos [aut]
Maintainer: Manuela Ott <manuela.ott@uzh.ch>
Depends: bayesmeta
License: GPL (>=2)
Manuela Ott, Malgorzata Roos Maintainer: Manuela Ott <manuela.ott@uzh.ch>
Ott, M., Hunanyan, S., Held, L., Roos, M. Sensitivity quantification in Bayesian meta-analysis. Manuscript revised for Research Synthesis Methods. 2020.
Spiegelhalter, D., Abrams, K., Myles, J. (2004). Bayesian Approaches to Clinical Trials and Health-Care Evaluation. John Wiley & Sons, Ltd.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 | # Acute Graft rejection (AGR) data analyzed in Friede et al. (2017),
# Sect. 3.2, URL: https://doi.org/10.1002/bimj.201500236
df <- data.frame(y = c(-2.310, -1.258), # log-odds-ratio
sigma = c(0.599, 0.642), # SE(log-odds-ratio)
labels = c(1:2))
# compute posterior, sensitivity and learning estimates for AGR data
# warning: it takes ca. 5-10 minutes to run this function
# on the above data set!
sensitivity_learning_table(df)
# dynamic prior adjustement based on RLMC
pri_par_adjust_dynamic(df=df, rlmc=0.25)
# static 5 % prior tail adjustement with reference threshold UU=1
pri_par_adjust_static(UU=1)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.