Intel® Summary Statistics Library

Submit New Article

Last Modified On :   May 19, 2008 10:16 AM PDT
Rate
 


What If Home | Product Overview | Intel® TM ABI specification | Technical Requirements
FAQ | Primary Technology Contacts | Discussion Forum | Blog

Intel® Summary Statistics Library 1.0 Update is available for downloading.

It includes several benefits and features:

Algorithm for parameterization of correlation matrix. The algorithm transforms the input which lacks property of positive semidefiniteness into the output meeting properties of correlation matrix. The algorithm is based on spectral decomposition method and can be used in financial computations.

Algorithm for computation of quantiles for streaming data. Computation of quantiles is done with pre-defined deterministic error and is highly efficient in terms of memory and speed. The algorithm can be used for processing of datasets which arrive in blocks.

Optimized algorithm for detection of outliers in datasets. The algorithm effectively utilizes all available cores of a multi-core system.

Dynamic libraries for IA-32/Intel® 64 Windows* and Linux* based platforms. Intel® Summary Statistics Library now provides greater flexibility of linking your application.

Bug fixes and other improvements. Incorrect processing of data in chunks for case of moments/covariance estimators and incorrect memory operations are fixed in the Update. Support of the case when most of the observations have zero weights is integrated into Intel® Summary Statistics Library. For greater flexibility rejection level alpha/n is replaced with alpha in the algorithm for detection of outliers. Set of examples that demonstrate use of the library is extended.

Known limitations. Intel® Summary Statistics Library algorithms for computation of quantiles (including streaming data case) should be used for task dimension 1.

 

Product Overview

Intel® Summary Statistics Library is a set of algorithms for parallel processing of multi-dimensional datasets. It contains functions for initial analysis of raw data which allow investigating structure of datasets and get their basic characteristics, estimates, and internal dependencies.

 

Features and Benefits

The library provides rich set of tools for estimation of various statistical characteristics of a dataset:

Basic statistics. Algebraic and central moments up to 4th order, skewness, kurtosis, variation coefficient, quantiles and order statistics.

Estimation of Dependencies. Variance-covariance/correlation matrix, partial variance-covariance/correlation matrix, pooled/group variance-covariance/correlation matrix.

Data with Outliers. The Intel® Summary Statistics Library contains a tool for detection of outliers in a dataset. Also the library allows computing robust estimates of the covariance matrix and mean in presence of outliers.

Missing Values. Data which contains missing values can be effectively processed using modern algorithms implemented in the package.

Out-of-Memory Datasets. Many algorithms of the library support data which cannot fit into the physical memory processing huge data arrays in portions. Specifically, variance-covariance matrix estimators, algebraic and central moments, skewness, kurtosis, and variation coefficient can process a dataset in portions.

Various Data Storage Formats. The Intel® Summary Statistics Library supports in-rows and in-columns storage formats for datasets, full and packed format for variance-covariance matrix.

The Intel® Summary Statistics Library uses recent advances of statistics by providing modern algorithms that enhance accuracy and performance of statistical computations. The library is optimized for latest multi-core Intel processors what allows to achieve significant performance benefits compared to traditional statistical packages and libraries.

 

Feedback


We are constantly striving to give you the most effective development tools for your business.  To help us shape our tools to your needs, please take a minute to answer 12 short questions. Your feedback is critical to the success of our product and will help drive the future direction of the Intel® Math Kernel Library.

 

Intel® Summary Statistics Library Blogs



Technical Requirements

 

1. To use Intel® Summary Statistics Library you must have a license for Intel® MKL product on your system. If you don’t, you can acquire the commercial product or try an evaluation copy.

2. Please see Install Guide for more details on technical requirements, including the list of supported processors and operating systems.

 

Frequently Asked Questions

 

Q - What programming interfaces does Intel® Summary Statistics Library support?
A - C and Fortran90/95

Q - Do I need additional software to use Intel® Summary Statistics Library?
A - Intel® MKL library is necessary to use Intel® Summary Statistics Library

Q - What do I need to get started for developing application with the Intel® Summary Statistics Library?
A - Download the library. The installation package includes full documentation which contains Installation Guide, User’s Guide, and User’s Manual.

Q - Where can I get support for the Intel® Summary Statistics Library?
A - You are welcome to join our WhatIf forum and post your questions and feedbacks.

Q - What are the licensing terms?
A - The licensing terms are listed on the download page.

 

Developer Support Team

 

Andrey Nikolaev
Andrey Nikolaev is a Senior Software Engineer at Software & Solutions Group. He holds primary degree in Applied Mathematics from Lomonosov Moscow State University, branch in city of Ulyanovsk and Ph.D. degree in Mathematical Cybernetics from Ulyanovsk State University. His work is related to design, development and optimization of statistical algorithms, optimization of financial algorithms, modeling and data analysis. Prior joining Intel he was involved in design of real-time SW for communication system in scientific industrial company.

Ilya Burylov
Ilya Burylov is a Senior Software Engineer at Software & Solutions Group. He holds a Master’s degree in Applied Mathematics from Perm State Technical University. Since joining Intel Ilya works on various problems related to scientific computation, design and development of sequential and parallel numerical algorithms, optimization of computational algorithms of Financial Mathematics.

Dmitry Kabaev
Dmitry Kabaev is a Senior Software Engineer at Software & Solutions Group. He holds a radio-physicist’s degree from Nizhniy Novgorod State University and a Ph.D. degree in Telecommunication from Federal Unitary Enterprise ‘Polyot’ (Russia). Since joining Intel his work is related to statistical data analysis, development and optimization of statistical numerical algorithms.