Changes in version 1.4.0 (2024-03-19) Changes - packages 'Synth', 'LowRankQP' and 'Rmalschains' have reappeared on CRAN. All references to the documentation of these packages, the support for 'malschains' as outer optimizer and the support of 'LowRankQPOpt' as inner optimizer have been restored. - some cryptic error messages are replaced by more meaningful ones, especially if necessary data is missing. - leaving character variables in the data no longer leads to an error. Bug fixes - corrected some missing edits (after copying and pasting) in one of the vignettes. - the package documentation has been corrected. Changes in version 1.3.9 (2023-11-29) Bug fixes - fixed insecure REprintf() calls in package's C code (many thanks to Kurt Hornik for the hint). Changes in version 1.3.8 (2023-11-13) Changes - removed support for package 'hydroPSO' (which has been archived). Bug fixes - fixed missing alias for package documentation (many thanks to Kurt Hornik for the hint). Changes in version 1.3.7 (2023-04-17) Changes - packages 'Synth', 'LowRankQP' and 'Rmalschains' have been archived recently. All references to these packages' documentation, the support for 'malschains' as outer optimizer and the support of 'LowRankQPOpt' as inner optimizer have been removed. - the vignettes now use cached copies of the results obtained with package 'Synth' (version: 1.1-6) because 'Synth' is currently not available on CRAN. some examples using package 'Synth' have been changed or removed. Changes in version 1.3.6 (2023-03-09) Bug fixes - corrected S3 generic/method inconsistency in ggplot.mscmt. Code relying on ggplot.mscmt's former argument order or the name of its first argument may need revision. Changes in version 1.3.5 (2023-01-31) New features - added support for weekly and daily data Bug fixes - removed (dispensable) LazyData entry from DESCRIPTION Changes in version 1.3.4 (2019-11-14) New features - added preliminary support for specification search (for the optimal set of predictors to include in the model specification) as an enhancement for cross validation Changes in version 1.3.4 (2019-11-14) - tolerance of lp solver changed slightly - added some flexibility (for the labels) to the ggplot method - removed some unused (and unexported) code - improveSynth now also checks whether the donor weights of synth()'s solution sum up to 1 (which they often do not!) and now also works when custom.v is set in the preceding synth()-call. Checks for optimality are now (solely) based on relative suboptimality. This leads to a slight change in the corresponding vignette. Bug fixes - added missing author information for LINPACK/SLATEC Fortran code (many thanks to Prof. Brian D. Ripley for the hint) Changes in version 1.3.3 (2018-05-04) Changes - minor documentation update (references) Bug fixes - Fixed a bug in mscmt for multivariate applications introduced in version 1.3.2 Changes in version 1.3.2 (2018-02-19) New features - added list element 'predictor.table' containing aggregated statistics of predictor values to output of 'mscmt' (feature request) Changes - list element 'dataprep.scaled' (in output of 'mscmt') now contains unscaled variants 'X0.unscaled' and 'X1.unscaled' of (aggregated) predictor values Changes in version 1.3.1 (2018-01-04) Changes - minor documentation update (references) Bug fixes - some fixes/increased robustness for edge cases (e.g. single predictor) - progress bar fixed (thereby eliminating compiler warnings) Changes in version 1.3.0 (2017-08-17) New features - cross validation is now supported by function mscmt, see the corresponding documentation. Changes - point size is now settable in ggplot.mscmt. - lp's in ill-conditioned problems are now tackled with the help of various lp solvers to improve robustness. Packages Rglpk and lpSolveAPI are required now. Bug fixes - 'no sunny donors' case fixed. - gamma-factors fixed. - some small bug fixes in ggplot.mscmt. Changes in version 1.2.0 (2017-01-30) New features - automatic repeated estimation with different outer optimizers and random seeds, optionally on a cluster. Changes - reasonable default parameters and documentation for outer optimizers have been added. - improved cluster support: (malfunctioning) load balancing with parLapplyLB replaced by load balancing via clusterApplyLB. Bug fixes - fixed special treatment of 'one sunny donor' case. Changes in version 1.1.0 (2016-11-29) New features - newly supported outer optimizer "none" to implant 'artificial' solutions (fixed w and optionally fixed v). - many new outer optimizers are now supported, more documentation will follow. - function compare() for comparison of estimation results, including corresponding updates for print.mscmt and ggplot.mscmt. Changes - names of agg.fns (component of mscmt results) are now set automatically. - user-defined agg.fns are now automatically exported to the cluster. - changed default optimizer to DEoptC. - predictor weights are not printed any more as verbose output in estimations. - some packages were moved from "Imports" to "Suggests". - number of calls to inner optimizer is now counted for benchmarking purposes. - package startup message added. Bug fixes - objective function weights alpha/beta/gamma are now treated correctly. - calls to the lp solver are now more robust. - some typos in vignettes have been corrected. - all data had to be sorted chronologically, this is now optional. - improveSynth reported wrong 'True' W*(V) if W*(V) was not optimal. Changes in version 1.0.0 (2016-07-25) New features - first public release of this package. Changes - no changes yet Bug fixes - no bugfixes yet