NAGNews 102 - 15 December 2011

Featuring


Copulas: a NAG Library Spotlight


In statistics, a copula can be thought of as defining the correlation structure for a family of multivariate distributions. Each distribution in the family is constructed by "gluing" together two or more univariate distributions, with the copula supplying the "glue". Copulas can be applied to a wide range of simulation problems.

copula

In Mark 23 of the NAG Fortran Library there are eight copulas including the Gaussian, Clayton/Cook-Johnson, Frank and Student's t copulas. Read more details here.


Examples of LabVIEW links to the NAG Library


We have just released a new collection of examples which illustrate the use of the NAG Library in the LabVIEW programming environment. They cover a variety of topics, including optimization, wavelet transforms, interpolation, statistical calculations and matrix operations, and make use of the NAG C Library, the NAG Fortran Library and the NAG Library for .NET. To try these new examples please go to the LabVIEW section of the NAG website. We are interested in receiving feedback from LabVIEW users about these examples, in order that they may be further refined and made more effective.


Update to R package provide access to more NAG routines


We have recently updated our R package (called NAGFWrappers) which enables R users to access to the Optimization chapters (E04 and E05) of Mark 23 of the NAG Fortran Library. The package also contains a few nearest correlation matrix solvers and some routines from the special functions chapter to give examples of the handling of parameter types (such as complex numbers).

The package includes documentation in the so-called Rdoc style that will automatically produce help pages within the R system.

It has been built and tested on 64 bit Linux, 32 bit Windows and 64 bit Windows, and can be downloaded from this page. We look forward to hearing your comments about its utility, as we are keen to further extend the package to make it still more useful to the R community.


NAG is awarded quality standard


The Numerical Algorithms Group has been awarded the international quality standard ISO 9001 certification.

The external assessors studied NAG's software development and peer review processes and NAG's investments in people and training that together enable NAG to provide regular product updates and an efficient service to its customers and partners.


Functions of Matrices, Nearest Correlation Matrix and Multi-Regime Factor Analysis


NAG Quant Event in New York City

The slides from the talks can be downloaded via the links below

Nicholas J Higham – The University of Manchester
Functions of Matrices and Nearest Correlation Matrices

Functions of matrices, which date back to the 1858 paper by Cayley that introduced matrix algebra, are of growing interest in many application areas due to the succinct and insightful way they allow problems to be formulated and solutions to be expressed. They are a valuable part of the problem solver's toolbox, as illustrated with practical examples relevant to finance. Nicholas Higham outlines the main matrix functions of interest and indicates the state of the art in methods for computing them. He also describes the University of Manchester's collaboration with NAG to extend the repertoire of matrix function codes in the NAG Library.

In many practical applications involving statistical modeling it is required to adjust an approximate, empirically obtained correlation matrix so that it has the three required properties of symmetry, positive semidefiniteness, and unit diagonal. He gives an overview of existing methods for computing the nearest correlation matrix to a given matrix and discuss an extension of the problem that incorporates factor structure, also indicating available software.

Paul G Hipes – Hipes Research, Inc
Multi-Regime Factor Analysis

Normal linear factor models provide an intuitive and common first step in the modeling of strongly correlated assets, but the assumption of normality is unrealistic for most financial market distributions, which typically display fat tails and skew. In order to improve their distributional accuracy while retaining their simplicity, we have generalized a class of factor models by permitting multiple regimes, thereby adding the parametric flexibility to capture fat tails and skew, while keeping the useful concept of factor loadings. Using maximum likelihood inference, we have estimated our multi-regime factor model with USD interest rate swap data. The regime-dependent factor loadings are consistent with an intuitive picture of multi-regime market dynamics. In addition, the posterior probabilities provide estimates of the time-series of regime residency. I will discuss these results and our general observations concerning the choice of the number of regimes, the influence of the regime-switching assumptions on the estimates, and the application of the model to risk management.

Alternatively you can go directly to the event webpage to download.


NAG in the news


Computational support for supercomputer service passes 50 person years

Fifty person years of expert skills have been allocated by NAG for computational science and engineering projects associated with the UK National HPC Service.

Building a better nuclear reactor with NAG

Researchers at Queen Mary, University of London and STFC Daresbury Laboratory (DL) who are employing the HECToR supercomputer to investigate radiation damage in containers have been making use of NAG's expertise in HPC software engineering.


iPad winner


We have received many positive suggestions in response to the question about why users benefit from the NAG Library and NAG services. Many of the replies reinforced the importance of NAG's reliability and the detailed documentation. The following are typical of comments received. … 'outstanding technology and reliable software'; 'quality, dedication and experience'; 'trustworthy', 'powerful yet easy'.

The iPad will be going to Joseph McDermott who suggested - 'Nobody knows About numbers like these Guys'.
(Note: Feedback closed on the 1st of December)


Recent Blog Posts


Keep up to date with NAG's recent blog posts here:

Coffee and Filters
Calling NAG routines from R


Out & About with NAG


Joint Mathematics Meeting, Boston, MA. 4-7 January 2012

HECToR Training Courses by NAG's HPC Team

Here are the courses being held over the next few months. Details of how to attend can be found on the HECToR website.

January 5-6, 2012
Imperial College London: OpenMP

January 9-11, 2012
Imperial College London: Parallel: Programming with MPI

January 18, 2012
University of Manchester: An Introduction fo CUDA Programming, Demystifying OpenMP

January 24-25, 2012
University of Leicester: An Introduction to CUDA Programming

February 8-10, 2012
University of Sheffield: Parallel Programming with

These HPC training courses are provided free of charge to HECToR users and UK academics.

For more information on any of the above events visit NAG's ‘Out & About’ webpage


NAG product implementations


The NAG Fortran Library, Mark 23 is now also available for the following platforms:

  • Linux64 (Intel 64/AMD64) using gfortran Compiler
    (NAG Product Code: FLL6A23DFL)
  • Apple Mac (Mac OS X) 64-bit systems using Intel Fortran Compiler
    (NAG Product Code: FLMI623DCL)

For full details of all available implementations, visit www.[token custom token_custom_nag_site]. Comprehensive technical details of each implementation are given in the relevant Installation and User Notes here.


NAGNews - Past Issues


We provide an online archive of past issues of NAGNews. For editions prior to these, please contact us.