고급 수치해석과 가시화 소프트웨어의 세계적 선두 주자로써 38년간의 경험과 실적을 갖고있는 비쥬얼뉴메릭스는 .NET용 수치해석 응용프로그램 개발 기술과 고급 차팅, 그래프 가시화에 대한 고객들의 증가하는 요구를 만족시키기 위해 IMSL™ C# Numerical Library의 새버젼 6.0을 출시하였습니다.

100% Microsoft Visual C# .NET으로 만들어진 유일한 수학, 통계, 챠팅 라이브러리 IMSL C# Library는 이익 극대화, 제품 디자인 최적화, 공급 체인 효율 최적화, 수요 예측등과 같은 일반적인 분석의 근간이 되는 강력한 알고리즘을 포함하고 있습니다.

Visual C# .NET 이나 Visual Basic .NET을 사용하는 개발자들은 빠르고 쉽게 어플리케이션에 내장할 수 있는 이미 입증된 IMSL 코드를 사용해서 신뢰성 있는 고성능의 데이터 분석을 할 수 있습니다.

※ IMSL C# Library 6.0 에서 새로 선보인 주요한 기능

- 새로운 차트 함수 : Tree Map 

- 새로운 데이터 마이닝 함수 :

• 나이브 베이즈 분류(Naive Bayes Classification) :분류 문제해결과 텍스트 마이닝 

- 새로운 시계열 분석 함수 :

• Auto_ARIMA : 시계열 분석을위한 ARIMA 매개변수 자동 예측 루틴
• Outlier Identification
• Lack-of-Fit Test  

- 생존분석 함수

• Life Table
• Kaplan-Meier 추정
• Cox's Proportional Hazards 

- 새롭게 추가된 통계 함수:

• 확률분포함수: Logistic , Parero, Non-central 분포함수들
• 난수생성 : Copula 분포 , Ziggurat 방법

- 새롭게 추가된 수학 함수

• Feynman-Kac PDE :블랙-숄즈 방정식(Black-Scholes Problems) 계산 함수
• Adams-Gear ODE
• Kochanek-Bartels Cubic Spline fitting함수
• Jacobian 계산을위한 Numerical Derivatives 

IMSL C# Library 은 세계적으로 가장 널리 사용되고 있는 IMSL 의 알고리즘을 이용해 수학, 통계 기능을 제공하므로 사용자는 블럭을 쌓는 것처럼 다양한 기능들을 조합하여 필요한 응용프로그램을 개발할 수 있으므로 코드를 만드는 시간과 비용을 줄일 수 있습니다.

또한, 제공되는 기능들은 사용자들이 원하는 어떠한 분야든지 적용할 수 있도록 만들어져 있으므로 IMSL 응용프로그램은 효과적이고 순수한 프로그래밍을 간단하고 완벽하게 제공할 것입니다.

< 영문 설명 >

Mathematical, Statistical and Charting Functionality

The algorithms available within the IMSL C# Library cover all of the major categories of functionality commonly used in numerical analysis, from commonly used math and statistical analysis functions like optimization and regression to advanced neural network and classification technology. By leveraging its pre-built algorithms, users can save weeks or months of development effort by embedding IMSL C# Library functions rather than building new algorithms from scratch.

This math and statistical algorithm functionality can be applied to an unlimited set of applications, such as bioinformatics and life sciences, fraud detection, risk management and portfolio optimization, manufacturing yield analysis and more.

※ High Performance Business Analytics Made Easy

The IMSL C# Numerical Library for Microsoft® .NET Applications is a numerical analysis library written in 100% C#, providing broad coverage of advanced mathematics and statistics for the .NET Framework.

Developers writing in C# or Visual Basic™ .NET (VB.NET) get seamless accessibility to analytics capabilities in the most integrated language for the .NET environment with the highest degree of programming productivity and ease of use with Visual Studio™.

The IMSL C# Library is the only numerical library of its kind to offer industry standard numerical analysis and charting for the .NET framework. This Library provides unprecedented analytic capabilities and the most comprehensive and accessible mathematical, statistical and financial algorithms for .NET languages. With the IMSL C# Library, Visual Numerics leveraged all of the benefits inherent in the .NET development framework by adding robust analytics to its broad set of capabilities.

※ Mathematical, Statistical and Charting Functionality

The algorithms available within the IMSL C# Library cover all of the major categories of functionality commonly used in numerical analysis, from commonly used mathematical and statistical analysis functions such as optimization and regression to advanced neural network and classification technology. By leveraging its pre-built algorithms, users can save weeks or months of development effort by embedding IMSL C# Library functions rather than building new algorithms from scratch.

This functionality can be applied to an unlimited set of applications, such as bioinformatics and life sciences, fraud detection, risk management and portfolio optimization, manufacturing yield analysis and more.

※ Advanced Analytics for Microsoft® .NET Applications®

※ Improved Performance

With V6.0, IMSL C# Library developers continue to have the option to use a version of the
library that embeds the Intel Math Kernel Library (MKL). The number of classes that utilize
MKL has been expanded resulting in improved application performance for many tasks
including linear algebra and matrix multiply functions.

※ Powerful Charting Capabilities

The IMSL C# Numerical Library V6.0 offers a new Tree Map class adding to the extensive array of charting types. The chart types range from traditional scatter plots and bar graphs, to statistical process control charts and box plots, to unique types such as dendrogram for hierarchical cluster analysis visualization.

Following a flexible object-oriented design, the customizable built-in types are easily embedded in desktop and web applications on Visual Numerics’ technical expertise and

※ New Functions

IMSL C# version 6.0 has added 27 new classes to its libraries providing unique numerical analysis techniques to customers solving analytic problems in finance, business intelligence, data mining and other areas of business, science and research. With version 6.0:

• Finance customers will benefit from a new Feynman-Kac algorithm that solves Black-Scholes problems, two new least squares optimizers, Random Copula methods, Survival Analysis, Logistic and Pareto distributions
• Business intelligence and data mining software developers can leverage Naive Bayes for classification and text mining problems. In addition, new classification capabilities in the Neural Network algorithm and new ways to select ARIMA models offer additional classification and forecasting techniques.
  

※ Typical Application Areas

Today, major corporations, academic institutions, and research laboratories worldwide use the IMSL C# Library. Example application areas include:

• Risk management in financial services and insurance
• Portfolio optimization in financial services
• Business intelligence extensions for data warehouse software
• Statistical analysis of manufacturing test data
• Inventory management and demand forecasting
• Medical and biological system R&D and modeling
  

※ Expert Consulting Services

The highly-skilled technical experts in Visual Numerics’ consulting organization collaborate with customers to identify specific application requirements at the initial phase of every project.

Visual Numerics’ consultants provide all levels of support from custom algorithm development to simply helping customers better understand their analysis and visualization needs.

Customers can rely on Visual Numerics’ technical expertise and dedicated hands-on help to achieve the highest return on investment.

※ New class that solves the generalized Feynman-Kac PDE and Black-Scholes problems

• Solves a generalized version of the Feynman-Kac partial differential equation that can be used in many financial modeling applications, including the Black-Scholes models with European or American style exercise opportunities on Calls or Puts. In the case of the Black-Scholes model these functions include many of The Greeks.
• Two white papers are available that provide more detail on this algorithm:
  • "Integrating Feynman-Kac Equations Using Hermite Quintic Finite Elements" is a technical white paper that describes the method for solving the Feynman-Kac PDE
  • "Solving Constrained Differential-Algebraic Systems Using Projections" is another technical white paper that describes specific software additions to the Differential-Algebraic Equation (DAE) solver code

※ Auto_ARIMA

• An “expert system” time series forecasting pre-conditioners and engine for forecasting situations where algorithm automation is desirable.
• Useful for data containing seasonality, level shifts, and outliers
• Automatic selection of model parameters for straightforward usage

※ Survival Analysis functions

• Cox’s Proportional Hazards model
• Kaplan-Meier (or product-limit) survival probabilities
• Kaplan-Meier reliability function estimates
• Population and cohort life tables

※ Many other new functions, including:

• Naive Bayes classification algorithm
• Random Copula methods and distributions
• Two new linear least-squares optimization classes
• Kochanek-Bartels Cubic Splines
• Logistic and Pareto CDFs and PDFs
• Several Non-central CDF and PDF distributions
• Treemap chart type

※ Enhancements to many existing algorithms, including:

• Improved algorithm for finding zeros of a function
• Faster normal random number generation
• Additional 2D tensor-product spline algorithm
• Additional ODE algorithm for stiff initial-value problems