In this paper, we introduce a new class of fuzzy linear regression model wi
th fuzzy input-fuzzy output data using shape preserving arithmetic operatio
ns on L-R fuzzy numbers for least-squares fitting. A mixed nonlinear progra
mming approach is used to derive the satisfying solution. This answers one
of Diamond's questions [Inform. Sci. 46 (1988) 141]. (C) 2001 Elsevier Scie
nce Inc. All rights reserved.