WSEAS Transactions on Mathematics, cilt.5, sa.6, ss.706-712, 2006 (Scopus)
A major problems in the modeling for econometricians as well as statisticians in the 21st century are the issue of uncertainty and ambiguity. In this study, we will focus in the context of fuzzy membership function of a sample selection model to deal with uncertainty and ambiguity of data. Fuzzy sets theory and its properties through the concept of fuzzy environment provide an ideal framework in order to solve the problem of uncertainty data. The purpose of this paper is to introduce fuzzy boundaries of a sample selection model that can be used to deal with sample selection model problems in which historical data contains some uncertainty. Instead of using crisp value i.e., single-valued parameters on which statistical inference may be done in the case of classical sample selection model, the quadratic term in term of membership function are used. The data of Malaysian Family and Population Survey 1994 are presented to shown the fuzzy boundaries of a sample selection models.