This is a suite of programs to calculate statistical properties of datasets.
Where possible, the programs can operate on arbitrarily large sets of data
without the need to store them internally: the data can be read point-by-point
and processed (e.g. to produce a running mean) such that the full set is
stored as data files and not as internal data arrays.
The programs are written in C++ using
the mathematical library and include:
MEANVARSD: running mean, variance and
standard deviation.
MOMENT: n'th moment of a dataset.
FRACTAL_DIMENSION: estimate fractal
dimensions of a distribution.
ENTROPY: calculate the entropy of a distribution.
CORRELATION: calculate correlations
(density matrix) between sets of time-series data.
ULTRAMETRIC: split time-series data
into subdominant ultrametric groups.
HEDGE: calculate odds to offer on a set
of probabilistic outcomes.