Software Package for Data Exploration, Univariate Analysis, Multivariate Analysis and Time Series Analysis
The statistical package Brodgar allows the user to carry out a range of statistical techniques in a Click-and-Go style. The package has an interface to R but also contains a couple of statistical methods not available in R, e.g. dynamic factor analysis, MAFA, chronological clustering.
The following techniques are available in Brodgar
- Data exploration: Boxplots, (multi-panel) Cleveland dotplots, (multi-panel) histograms, (multi-panel) scatterplots, pairplots, design and interaction plots.
- Univariate analysis: Regression, GLM, GAM, mixed modelling, GLS, univariate regression trees. Power analysis
- Multivariate analysis: Principal component analysis, correspondence analysis, redundancy analysis, canonical correspondence analysis, MDS, NMDS, multivariate regression trees, ANOSIM, Mantel test, clustering techniques.
- Time series techniques: Dynamic factor analysis, MAFA, chronological clustering, auto-correlation functions.
Brodgar compliance statement on R GNU license
The software package R is distributed under the GNU general public license. Brodgar, which is not distributed under the GNU license, creates ascii files containing R script commands which are sourced into the binary version of R (using the BATCH mode). Although Brodgar is linked to R, it does not contain it. The user will need to (i) download a compiled version of R, (ii) install it and (iii) tell Brodgar where it can find R. Hence, R and Brodgar are two different packages (Brodgar is the shell and R the compiler) which are an arm length apart. As a consequence, Brodgar complies with the GNU license. See http://www.gnu.org/copyleft/gpl.html for details.