CRAN 0.4.0

fix in plotBoxplot function after ggplot2 update

Data and exports cleaning

• create a lighter version of Pisa dataset
• make some plotting functions public

Weighted models and models with covariates

New functionalities:

• weighted models,
• models with covariates,
• enabled 'by-formula' model passing in the mergeFactors function,
• new tests.

Faster clustering

Parameters method and subsequent were merged together. Possible values for method are:

• fast-fixed (method = "hclust" + successive = TRUE),
• fixed (method = "hclust" + successive = FALSE),
• fast-adapive (method = "LRT" + successive = TRUE),
• adaptive (method = "LRT" + successive = FALSE).

Some improvements in algorithms and in plots alignment were done.

CRAN 0.3.1

As appeared on CRAN (https://cran.r-project.org/web/packages/factorMerger/index.html)

New visualizations and algorithm speedup

Introduced new functionalities:

• Tukey post hoc visualization.

Additional updates:

• multi dimensional Gaussian loglik calculation speedup,
• changes in single dimensional Gaussian plot (right panel -- means and their 95% conf. interval)

factorMerger v0.2

Introduced new functionalities:

• overloaded function plot.factorMerger,
• predicting new labels for factor groups with cutTree

factorMerger v0.1.1

Set of tools to support results from post hoc testing for parametric models.

Models supported

• one-dimensional Gaussian (with the argument family = "gaussian"),
• multi-dimensional Gaussian (with the argument family = "gaussian"),
• binomial (with the argument family = "binomial"),
• survival (with the argument family = "survival").

Hypothesis testing

• all-to-all (with the argument successive = FALSE),
• successive (with the argument successive = TRUE).
  The version all-to-all considers all possible pairs of factor levels. In the successive approach factor levels are preliminarily sorted and then only consecutive groups are tested for means equality.

Merging stategies

• Likelihood Ratio Test (with the argument method = "LRT"),
• agglomerative clustering with constant distance matrix (with the argument method = hclust, based on the DMR4glm algorithm by Agnieszka Prochenka).