A least-absolutes approach to multi-dimentional fuzzy regression modellig is introduced and investigated for the case of crisp input/fuzzy output data, by using a generalized Hausdorff metric.

A comparative study, based on three well-known goodness-of-fit indices and three data set, including a real agricultural data set, indicate the performance of the proposed approach with respect to some common approaches to fuzzy regression modelling.

Keywords: Imprecise data; Fuzzy regression; Hausdorff metric; Least-absolutes method

Biography: S. Mahmoud Taheri, Born 1965

Title: Associate Professor

Address: School of Mathematical Sciences, Isfahan University of Technology, Isfahan 8415683111, Iran E.mail: Taheri@cc.iut.ac.ir

Study and Work Experience:

1984-1988: B.Sc., Dept. of Statistics, School of Sciences, Ferdowsi University, Mashhad, Iran

1988-1991: M.Sc., The same above

1995-2000: Ph.D., Dept. of Statistics, School of Sciences, Shiraz University, Shiraz, Iran

2000-2005: Assistante Professor, School of Mathematical Sciences, Isfahan University of Technology, Isfahan, Iran

2006-Present: Associate Professor, The same above

1999-2000: Visiting Scholar, Dept. of Statistics, University of British Columbia, Vancouver, Canada

2002-2002: Research Scholar, ICTP, Trieste, Italy

2004-2004: Research Scholar, ICTP, Trieste, Italy

2009-2010: Research Visiting (As Sabbatical), Dept. of Statistics, Wien University of Technology, Wien, Austria

Research Interest: Statistical Inference, Statistics and Probability with Fuzzy Information, Fuzzy Regression Analysis.

Publications: 22 International Journal Papers, 14 International Conference Articles, 2 (Persian) Books.