Rsolnp (version 1.11)

benchmark: The Rsolnp Benchmark Problems Suite.

Description

The function implements a set of benchmark problems against the MINOS solver of Murtagh and Saunders.

Usage

benchmark(id = "Powell")

Arguments

id

The name of the benchmark problem. A call to the function benchmarkids will return the available benchmark problems.

Value

• A data.frame containing the benchmark data. The description of the benchmark problem can be accessed throught the description attribute of the data.frame.

Details

The benchmarks were run on an intel core 2 extreme x9000 cpu with 8GB of memory on windows vista operating system. The MINOS solver was used via the tomlab interface to matlab except for the portfolio problems (Rachev and Kappa ratios) which used the SNOPT solver.

References

W.Hock and K.Schittkowski, Test Examples for Nonlinear Programming Codes, Lecture Notes in Economics and Mathematical Systems. Springer Verlag, 1981. Y.Ye, Interior algorithms for linear, quadratic, and linearly constrained non linear programming, PhD Thesis, Department of EES Stanford University, Stanford CA. B.A.Murtagh and M.A.Saunders, MINOS 5.5 User's Guide, Report SOL 83-20R, Systems Optimization Laboratory, Stanford University (revised July 1998). P. E. Gill, W. Murray, and M. A. Saunders, SNOPT An SQP algorithm for large-scale constrained optimization, SIAM J. Optim., 12 (2002), pp.979-1006.

Examples

benchmarkids()
benchmark(id = "Powell")
benchmark(id = "Alkylation")
benchmark(id = "Box")
benchmark(id = "KappaRatio")
benchmark(id = "RosenSuzuki")
benchmark(id = "Wright4")
benchmark(id = "Wright9")
benchmark(id = "Electron")
benchmark(id = "Permutation")
# accessing the description
test = benchmark(id = "Entropy")
attr(test, "description")