The Numerical Algorithms Group (NAG) has released an updated version of its NAG Library For Java with over 100 additional routines and improved error checking.

Now at  its second release the NAG Library for Java enables the calling of 1,784 mathematical and statistical routines to aid complex computation and now features enhanced error reporting enabling increased precision from computation results.

Release 2 also provides additional abstract classes for callback functions alongside the following new numerical functionality introduced at Mark 24 of the NAG Library

• Multi-start (global) optimization
• Non-negative least squares (local optimization)
• Nearest Correlation Matrix
• Inhomogeneous Time Series
• Gaussian Mixture Model
• Brownian Bridge & random fields
• Best subsets
• Real sparse eigenproblems
• Matrix Functions
• Two-stage spline approximation to scattered data
• Confluent Hypergeometric Function (pictured below)

The NAG Library offers detailed documentation giving background information and function specification and in addition guides users to the right function for their problem via decision trees.

To use the NAG Library you need to have installed a copy of the NAG Fortran Library, available in both  32 and 64-bit versions for Windows and Linux, and to download the NAG Library for Java wrappers. If you are not already a customer you can request a trial version from NAG Software Trials.