ABSTRACT
This paper presents a procedure that incorporates scatter search and threshold accepting to find the maximum likelihood estimates for an econometric model –– multinomial probit (MNP) model. A set of numerical tests, based on a synthetic data set with known model specification and error structure, were conducted to test the effectiveness and efficiency of the proposed framework. The results indicated that the proposed procedure can enhance performance in terms of likelihood function value and computation efficiency for MNP model estimation.
