Welcome to Intel® Summary Statistics Library, solution for parallel statistical processing of multi-dimensional datasets. It contains functions for initial statistical analysis of raw data which allow investigating structure of datasets and get their basic characteristics, estimates, and internal dependencies.
The library provides rich set of tools intended to compute various statistical estimates for datasets:
• It includes functions for computation of algebraic and central moments up to 4thorder, skewness, kurtosis, variation coefficient, quantiles and order statistics.
• The package contains collection of algorithms for estimation of dependencies in the dataset: variance-covariance/correlation matrix, partial variance-covariance/correlation matrix, pooled/group variance-covariance/correlation matrix. The library also allows computing robust estimates of covariance matrix and mean for data in presence of outliers.
• Outliers in datasets can be detected using the tool of Intel® Summary Statistics Library.
• Data which contains missing values can be effectively processed using algorithms of the library.
Algorithms of Intel® Summary Statistics Library provide support for data which are huge to be analyzed at once and are processed in chunks. In particular, basic statistics estimators like variance-covariance matrix, algebraic and central moments, skewness, kurtosis, and variation coefficient support datasets which are available in chunks.
Intel® Summary Statistics Library supports in-rows and in-columns storage formats for datasets, full and packed format for variance-covariance matrix.
Intel® Summary Statistics Library supports single and double precision floating point arithmetic.
Algorithms of the Intel® Summary Statistics Library are designed to take advantage of multi-core processors.
Intel® Summary Statistics Library API provides interfaces for FORTRAN 90 and C/89 languages.