The following simplified statistical functions have been added.
- CORRELATION, which returns the degree of correlation between two independent sets of data.
- KMEANS_CLUSTER, which partitions n observations into k clusters in which each
observation belongs to the cluster with the nearest mean.
- STDDEV, which quantifies the amount of variation or dispersion
of a set of data values.
- MULTIREGRESS, which derives a linear equation that best fits the data values of one or more numeric fields, and uses this
equation to create a new column in the report output.
- RSERVE, which runs an R script that returns vector output.
These functions can be called in a COMPUTE command and operate on the internal matrix that is generated during TABLE request
processing. The STDDEV and CORRELATION functions can also be called as a verb object in a display command.
For more information, see the Using Functions technical content.