The IMSL Fortran Numerical Library consists of two separate but coordinated Libraries that allow easy user access. These Libraries are organized as follows:

Most of the routines are available in both single and double precision versions. Many routines for linear solvers and eigensystems are also available for complex and complex-double precision arithmetic. The same user interface is found on the many hardware versions that span the range from personal computer to supercomputer.

This library is the result of a merging of the products: IMSL Fortran Numerical Libraries and IMSL Fortran 90 Library.

The IMSL Fortran Numerical Library contains functions which may take advantage of functions in vendor supplied libraries such as Intel’s® Math Kernel Library (MKL) or Sun’s High Performance Library. Functions in the vendor supplied libraries are finely tuned for performance to take full advantage of the environment for which they are supplied. For these functions, the user of the IMSL Fortran Numerical Library has the option of linking to code which is based on either the IMSL legacy functions or the functions in the vendor supplied library. The following icon in the function documentation alerts the reader when this is the case:

Details on linking to the appropriate IMSL Library and alternate vendor supplied libraries are explained in the online README file of the product distribution.

The IMSL STAT LIBRARY is a collection of FORTRAN subroutines and functions useful in research and statistical analysis. Each routine is designed and documented to be used in research activities as well as by technical specialists.

To use any of these routines, you must write a program in FORTRAN (or possibly some other language) to call the STAT LIBRARY routine. Each routine conforms to established conventions in programming and documentation. We give first priority in development to efficient algorithms, clear documentation, and accurate results. The uniform design of the routines makes it easy to use more than one routine in a given application. Also, you will find that the design consistency enables you to apply your experience with one STAT LIBRARY routine to all other IMSL routines that you use.

The STAT LIBRARY is organized into chapters; each chapter contains routines with similar computational or analytical capabilities. To locate the right routine for a given problem, you may use either the table of contents located in each chapter introduction, or one of the indexes at the end of this manual.

Often the quickest way to use the STAT LIBRARY is to find an example similar to your problem and then to mimic the example. Each routine document has at least one example demonstrating its application. The example for a routine may be created simply for illustration, it may be from a textbook (with reference to the source) or it may be from the statistical literature, in which case IMSL routine GDATA retrieves the data set.