MVSP 多變數統計分析軟體

MVSP is an inexpensive and easy to use program that performs a number of multivariate numerical analyses useful in many scientific fields. It calculates three basic types of eigenanalysis ordinations: principal components, principal coordinates, and correspondence/detrended correspondence analyses.

Version 3 adds canonical correspondence analysis, a technique highly popular in ecological studies. It can also perform cluster analysis, with 23 different distance and similarity measures and seven clustering strategies.

MVSP is in use at hundreds of sites in 86 countries and its results have been published in numerous scientific journals, including Science and Nature.

MVSP performs several types of eigenanalysis ordinations: principal components analysis (PCA), principal coordinates analysis (PCO), and correspondence/detrended correspondence analysis (CA/DCA). It also does canonical correspondence analysis (CCA), a technique highly popular in ecological studies.

It can also perform cluster analysis, using 23 different distance or similarity measures and seven clustering strategies. A dual clustering option lets you produce dendrograms of both variables and cases in one step, along with a copy of the original data matrix, sorted in the same order as the dendrograms. Constrained clustering may also be performed, so that the original input data order is maintained. Diversity indices may be calculated on ecological data; these include Simpson's, Shannon's, and Brillouin's indices.

The number of cases and variables that can be analyzed is limited only by the amount of memory available to Windows (RAM and hard disk swap file), up to a maximum of 2 billion cases and variables.