NAGnews 115 | 11 July 2013

In this issue


Turbo-charged numerics: the NAG Library for SMP & Multicore - now available


At the International Supercomputing Conference (ISC13), Leipzig in June, NAG launched Mark 24 of the NAG Library for SMP & Multicore.

The Library for SMP & Multicore now contains over 1,750 mathematical and statistical algorithms with over 130 new routines being added at this update. Exciting new functionality includes:

  • Multi-start (global) Optimization
  • Non-negative Least Squares
  • Nearest Correlation Matrix
  • Inhomogeneous Time Series (today's functionality spotlight)
  • Gaussian Mixture Model
  • Confluent Hypergeometric Function (1F1)
  • Brownian Bridge
  • Best Subsets
  • Real Sparse Eigenproblems
  • Matrix Functions
  • Two Stage Spline Approximation

If you're an existing, supported, NAG Library for SMP & Multicore user we urge you to upgrade to the latest Mark. Using the latest versions of our Libraries means you'll have access to all the exciting new functionality. We are happy to guide users through the upgrade if necessary. Contact us for more information.

Examples of Mark 24 Performance


New Functionality Spotlight (Mark 24): Inhomogeneous Time Series


We continue to shine the spotlight on the new functionality in Mark 24 of the NAG Library. Today we focus on the new 'Inhomogeneous Time Series'. Three new routines in the Time Series Analysis Chapter (G13) G13MEF, G13MFF and G13MGF.

The Time Series Analysis (G13) chapter of the NAG Library contains a wide range of techniques for investigating and modelling the statistical structure of univariate and multivariate time series, that is, a series of observations collected at different points in time. In this discussion we will denote a generic univariate time series as a sequence of pairs of values (zi; ti), for i = 1; 2; : : : ; where the z's represent an observed scalar value and the t's the time that the value was observed.

Prior to Mark 24, all of the methods included in (G13) required the time series to be homogeneous, that is, the sampling times had to be equally spaced, with ti - ti?1 = for all I and some value _. In many real-world applications this assumption does not hold; the observed series is inhomogeneous. Standard time series analysis techniques cannot be used on an inhomogeneous series without first preprocessing the series to construct an artificial homogeneous series by, for example, resampling the series at regular intervals. At Mark 24 we have introduced a series of operators, suggested by Zumbach and Muller [1], that can be used to extract robust information directly from an inhomogeneous time series. In this context, robust information means that the results should be essentially independent of minor changes to the sampling mechanism used when collecting the data, for example, changing a number of time stamps or adding or removing a few observations.

Read the full paper on Inhomogeneous Time Series (including an example) here.


NAG's HECToR Team Win HPC Innovation Excellence Awards


NAG is delighted to have been presented with two HPC Innovation Excellence Awards at the International Supercomputing Conference (ISC13) which was held last month in Leipzig. The awards, presented to the NAG HECToR Computational Science and Engineering (CSE) Team on behalf of IDC - global market intelligence company, and the HPC User Forum Steering Committee were for two of the many l application performance improvement projects successfully undertaken by NAG in their role of providing CSE support for the UK's national supercomputing service, HECToR.

The two key projects recognised by the awarding bodies were:

'Performance of Quantum Monte-Carlo Application (CASINO) Quadrupled' with the University College London; and

'Performance of Molecular Dynamics Application (DL_POLY_3) 20x Faster after Optimisations' with STFC and the University of Warwick.

NAG is delighted to have its HPC expert services recognised and acknowledged in this way. Speaking after receiving the award at ISC13, Dr Mike Dewar, Chief Technology Officer at NAG said "NAG HPC software performance experts have delivered major performance and functionality gains for many users of supercomputers over the last decade. The NAG HECToR Team in particular has helped the research users of HECToR to achieve substantial increases in the science and engineering output through over 50 successful application improvement projects. We are proud that the NAG HECToR team has been awarded the HPC Innovation Excellence Awards and will continue to serve our commercial and academic users around the world".

The full story is here.


Training Courses - Numerical Programming for C/C++ and Java with NAG


You are invited to attend two complimentary training workshops organised by the Numerical Algorithms Group (NAG) in Oxford on Thursday 25th July 2013. The courses are free to attend, but places are limited so do register as soon as possible. You are welcome to attend one or both workshops and lunch is included for all attendees.

Numerical Programming in C++ with the NAG C Library
(morning workshop)
Hosted by Chris Seymour

C++ is a prevalent language in the development of financial applications. This workshop will give attendees insight to using the NAG C Library with C++. The workshop will feature a hands-on session where NAG experts will guide you through calling the routines in the C Library into C++.

Course attendees will be using the latest version of the NAG C Library, Mark 23 in this workshop. New functionality added to the C Library at Mark 23 include: Matrix functions, Multi-start Optimization, Additions to Nearest Correlation Matrix, Bound Optimization BY Quadratic Approximation and Skipping Ahead the Mersenne Twister Random Number Generator.

Introduction to the NAG Library for Java
(afternoon workshop)
Hosted by Ludovic Henn

NAGs latest numerical library the NAG Library for Java provides Java developers with over 1,700 mathematical and statistical routines. This workshop will give an introduction to the new Library and demonstrate its ease of use. Attendees will have time to get hands-on with the Library.

Guarantee your place by registering here.


How to compute a definite integral over a finite range to a specified relative accuracy - NAG and Excel new example


We continue to work on examples showing the flexible nature of NAG routines. There are lots of NAG and Excel examples here. A new example has been recently published at the request of a user which demonstrates how to compute a definite integral over a finite range to a specified relative accuracy using a method described by Patterson. This method is implemented in the NAG Library as routine D01AHF (nagf_quad_1d_fin_well).

How to compute a definite integral over a finite range to a specified relative accuracy using a method described by Patterson. Using the NAG Fortran Library NEW EXAMPLE

 


Events & Training Courses


     
  • Actuarial Teachers' & Researchers Conference 2013
    18th-19th July 2013, Keele University
    This event provides those interested in actuarial research and education with the opportunity of sharing ideas and to catch up on the latest developments. This is the only actuarial conference with a research and education focus in the UK which is truly cross-practice and open to all. following on from last year's success, the conference will continue the theme of 'bridging the gaps' between academia and practitioners, between academic disciplines, and/or between actuaries and other professions.

    David Humphris and Martyn Byng from NAG are in attendance, and will be making a joint presentation on Thursday 18th July. Title: Numerical Software Developments Relevant to the Actuarial Community.
     
  • Advanced Risk & Portfolio Management Bootcamp
    12th-17th August 2013, New York
    NAG is sponsoring this 'bootcamp' course for risk and portfolio management practitioners.
     
  • 3rd IMA Conference on Mathematics in Defence
    Thursday 24 October 2013. Malvern.
    This conference brings together a wide variety of mathematical methods with defence and security applications. The conference programme will include keynote speakers, contributed presentations and poster sessions as well as refreshment breaks for informal discussions. It is intended for mathematicians, scientists and engineers from industry and academia, as well as government personnel who have an interest in how mathematics can be applied to defence problems. We will once again be present at our exhibition booth and look forward to meeting you if you are in attendance. Phil Ridley of NAG will also be presenting a paper entitled 'The CABARET Scheme for Scalable High-Resolution Shock Capturing in Turbulent Flows'.
     


Recent blog posts


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

A quick but dirty way to use the NAG C Library Mark 23 from Python

I get requests about using NAG from python from time to time, and NAG strives to make its algorithms as accessible as possible from a wide variety of languages and platforms. Python is no exception.

While we don't yet provide a fully documented python API replete with examples and so on, there is a quick but dirty way to get NAG C Library python bindings covering every library algorithm, struct, and enum.

NAG's John Morrissey has written about a new way of using the NAG Library on Python - read how it's done here - blog post by John Morrissey.

The NAG Library for Java - An Illustrative Example Using Black-Scholes Pricing

This blog contains an illustrative example of how to use the NAG Library for Java in which we use the problem of pricing a European option as a concrete problem to be solved. The problem is particularly simple, for which an explicit solution is known, however in the real world the problems are more complicated and alternative methods have to be employed. For this reason we use a very basic Monte-Carlo method to illustrate the use of the NAG random number generators, whilst acknowledging that in practice a number of variance-reduction techniques would employed by the professional.

Monte Carlo simulation

Monte Carlo simulation uses repeated random sampling to compute numerical results. It is used heavily in financial computation. In this article, I use quasi-random numbers (the NAG routine G05YJF) and pseudo-random numbers (the NAG routine G05SKF and the explicit Black-Scholes formula (the NAG routine S30AAF) to provide benchmark. Computation time and accuracy are measured to compare the performance of the two different random number generators.

Read the full post by Peter Guo.


NAGNews - Past Issues


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