LINPRO: Linear inverse problem library for data contaminated by statistical noise

The library LINPRO which provides the solution to the linear inverse problem for data contaminated by a statistical noise is presented. The library makes use of two methods: Maximum Entropy Method and Singular Value Decomposition. As an example it has been applied to perform an analytic continuation of the imaginary time propagator obtained within the Quantum Monte Carlo method.Program summaryProgram title: LINPRO v1.0.Catalogue identifier: AEMT_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AEMT_v1_0.htmlProgram obtainable from: CPC Program Library, Queen’s University, Belfast, N. Ireland.Licensing provisions: GNU Lesser General Public Licence.No. of lines in distributed program, including test data, etc.: 110620.No. of bytes in distributed program, including test data, etc.: 3208593.Distribution format: tar.gz.Programming language: C++.Computer: LINPRO library should compile on any computing system that has C++ compiler.Operating system: Linux or Unix.Classification: 4.9, 4.12, 4.13.External routines: OPT++: An Object-Oriented Nonlinear Optimization Library [1] (included in the distribution).Nature of problem: LINPRO library solves linear inverse problems with an arbitrary kernel and arbitrary external constraints imposed on the solution.Solution method: LINPRO library implements two complementary methods: Maximum Entropy Method and SVD method.Additional comments: Tested with compilers–GNU Compiler g++, Intel Compiler icpc.Running time: Problem dependent, ranging from seconds to hours. Each of the examples takes less than a minute to run.References:[1] OPT++: An Object-Oriented Nonlinear Optimization Library, https://software.sandia.gov/opt++/.