NEWS R Documentation

## News for Package ragt2ridges

### Changes in version 0.3.4

• The packages Biobase and CHGbase are now imported (instead of suggested) in the DESCRIPTION-file to avoid errors on CRAN.

### Changes in version 0.3.3

• Modification of input checks to meet CRAN's best practice.

• Citations are updated.

### Changes in version 0.3.2

• The packages Biobase and CHGbase are now suggested in the DESCRIPTION-file to avoid errors on CRAN.

• The .Rd-file of the package is updated.

• Citations are updated.

• A typo corrected.

• Some warnings on C++-code tackled.

### Changes in version 0.3.1

• The functions motifStatsVAR1 and pruneMotifStats added.

• Dependency on the marray-package removed (often give an error in CRAN checks). Consequently, the color scheme employed in the evaluateVAR1fit is less refined.

### Changes in version 0.3.0

• More extensive description added to the description field of the description file.

• Correct package name inserted in registerDynamicSymbol.c.

• R code aligned for nicer viewing on Github.

• The hpvP53mir data object is replaced by the correct version.

• Several typo's in manuals corrected.

• More consistent wording across manuals introduced.

• Citation updated.

• Functions momentS, ridgePchordal, optPchordal, and support4ridgep are now imported from the rags2ridges-package.

• Three new modules are added: fused ridge estimation of multiple VAR(1) models, and ridge estimation of the VAR(2) and VARX(1) models.

### Changes in version 0.2.4

• Bug repaired in the graphVAR1-function. Input combination type="globalPC" and prune=TRUE produced a warning.

• Some code from Rcpp-functions replaced by slightly more efficient code.

• Meaningless code from Rcpp-functions removed.

• Citation updated.

• More data from the same experiment have been added. Documentation modified accordingly. Of note: the original data have been renamed to hpvP53rna.

• A misnomer in the output and documentation of the nodeStatsVAR1-function fixed.

• The formula of the VAR(1) model is added to documentation of the ridgeVAR1-function.

### Changes in version 0.2.3

• Some re-organization of the src-directory.

• Rcpp-functions made exportable to other packages.

• Input checks of the ridgeVAR1-function contained inconsistencies: repaired.

• Function momentS added: evaluates the moments of the sample (inverse) covariance matrix.

### Changes in version 0.2.2

• Change to the description and namespace-file. Importing now from methods, for compatibility with Bioconductor 3.4.

• A bug repaired in the ridgeVAR1-function. The option-combination fitA="ss" and a non-empty matrix being passed to the argument zerosA gave an error (an argument was not supplied to an internal C++ function).

• A bug repaired in the ridgeVAR1-function. When specifying the option-combination fitA="ss" and non-empty matrices being passed to the arguments zerosA and zerosP the wrong order of arguments was supplied to an internal C++ function.

### Changes in version 0.2.1

• Extra argument (TburnIn) for dataVAR1-function. It specifies number of burn-in time points before data is sampled from the process.

• The impulseResponseVAR1-function and its arguments changed slightly. Previously, the function aggregated all impulse responses between a variate at the current time point and those at the T future time points. The current implemention evaluates the impulse response in each variate at the current time point t on those at time point t+T. Apologies for inconvenience caused.

• The mutualInfoVAR1-function and its arguments changed slightly. Previously, the function aggregated all mutual informations between a variate at the current time point and those at the T future time points. The current implemention evaluates the mutual information between each variate at the current time point t and those at time point t+T. Apologies for inconvenience caused.

• The nodeStatsVar1-function changed internally to accomodate the changes in the impulseResponseVAR1 and mutualInfoVAR1-functions.

• Extra argument (zeros) for the sparsifyVAR1-function. It specifies which elements in the auto-regression coefficient matrix are known to zero and should not be taken along in the sparsification.

### Changes in version 0.2.0

• Added a NEWS file.

• Modified the example code of the evaluateVAR1fit. In this version, the evaluateVAR1fit function is no longer executed in the example code. This avoids saving .eps files during the package checking.