Departmental Papers (ASC)
Document Type
Book Chapter
Date of this Version
1-1982
Publication Source
A General Survey of Systems Methodology
Start Page
299
Last Page
304
Abstract
Exploring multi-variate data involves searching for a simple model that simulates, explains or accounts for given data within tolerable margins of errors. Such a search requires procedures for generating suitable models that could be tested and compared. We are proposing here several rules for transforming the symbolic descriptions of a model which when inserted into a suitable algorithm will compute generations of structural models, one minimally simpler than the other. Although we are primarily concerned here with rules that generate models without loops for which algorithms are currently unavailable (Klir, 1980), we will contrast the transformation rules for this class of models with those responsible for generating all possible models and we will furthermore distinguish between rules that generate models with the same and with different covers.
Recommended Citation
Krippendorff, K. (1982). A proposal for an algorithm for generating loopless or recursive models of multi-variate data. In L. Troncale (Ed.), A general survey of systems methodology: Proceedings of the twenty-sixth annual meeting of the Society for General Systems Research with the American Association for the Advancement of Science, Washington, DC, January 5 - 9, 1982 (pp. 299-304). Louisville, KY: Society for General Systems Research. Retrieved from http://repository.upenn.edu/asc_papers/208
Date Posted: 06 October 2010