The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
rsDCM 0.2.0
Robust and sparse group
inference
- Replaced mode-only inclusion summaries with integrated posterior
inclusion probabilities estimated by a registered C++ parallel-tempering
sampler.
- Added batch-means Monte Carlo standard errors, chain-specific PIPs,
acceptance rates, swap rates, and maximum between-chain PIP ranges.
- Reworked mode fitting as a multistart generalized EM algorithm with
conditional maximization, monotonicity safeguards, stationarity checks,
and an
nlminb() fallback for covariance-component
optimization.
- Added conditional Laplace covariance matrices and standard
deviations for group coefficients.
- Added selectable precision-Cholesky, covariance-Cholesky, and
symmetric whitening implementations. Returned weights are explicitly
identified as whitening-coordinate diagnostics.
- Fixed the residual scale by default for scale identification and
added an opt-in free-scale mode with a warning.
- Added diagnostics for multistart convergence, objectives, Q-function
monotonicity, and fitted covariance components.
- Replaced the bundled preliminary NARPS summaries with corrected
within- subject PEB summaries for the same 48 participants.
Package organization
- Consolidated the novel method in
R/rsDCM.R.
- Consolidated MATLAB-ported DCM and PEB functions in
R/DCM.R.
- Moved the C++ sampler to the standard
src/ directory
with registered native routines.
- Updated manuals, examples, tests, citation information, and the
vignette.
rsDCM 0.1.0
Initial release.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.