Distributions f(x) measured in high-energy-physics experiment are often distorted or transformed by limited acceptance and finite resolution of the detector. The unfolding of measured distribution is an important, but due to the inherent instability a very difficult problem. A talk about unfolding has been given in the 1984 CERN Summer School for Computing in Aiguablava, Spain (CERN Yellow Report 85-02), and DESY 84-118 (Red Report).
The manual and the source code of the unfolding program RUN, written in Fortran77, is available below. Input to the unfolding program are ntuples for data and Monte Carlo quantities. The unfolded distribution f(x) is reconstructed in bins by a regularization method, which includes the determination of the covariance matrix by error propagation from the data statistics and the properties of the resolution function.
There is of course no guarantee that the program is working correctly. The program has been developed about 15 years ago for the use by the author. Its usage requires some understanding of unfolding and statistics in general, especially with respect to the optimization of the various parameters.
A new version, with automatic optimization of the various parameters,
is being developled, but will not be released.
There are 16 source files with Fortran programs. Some source files are specific for the RUN program, others contain utility routines (not all of the utility routines in the source file are used in RUN).
The software can be freely used for research and education. We expect that all publications describing work using this software quote at least one of the references below.
Disclaimer: This software is provided without any expressed or implied warranty. In particular thre is no warranty of any kind concerning the fitness of this software for any particular purpose.