Genstat 23 new features overview
Introducing Genstat 23. This short video highlights some of the new features in this latest release.
Censored Poisson count data
Genstat makes it easy to analyse censored Poisson count data using a log-linear model, a generalized linear mixed model, or a hierarchical generalized linear model. Left and right censoring options are available on each of these menus, allowing you to set upper or lower values beyond which data will not be counted.
2D Trellis plot of groups
2D Trellis plots of groups are graphs that enable you to explore very large data sets in which observations are classified into groups. This graph type lets you investigate the distribution of groups of observations over two dimensions, that is, two variates, using either a trellis of bar charts or a trellis of pie charts.
Mosaic plots are a very powerful tool for visualising and exploring complex tables of counts, enabling you to identify patterns and relationships between multiple categorical variables in a single plot. Within the plot, each group (i.e., cell in the table) is represented by a coloured box, where the size of the box is proportional to the number of observations in that group.
The aim of cluster analysis is to classify the observations into distinct groups based on how similar their data are. Genstat offers an efficient clustering algorithm specifically designed to handle very large multivariate data sets. This algorithm exploits principal components to reduce the dimensions of the data. Additionally, it uses a specialized clustering method to reduce the heavy computational load when dealing with a large number of observations.