Effective estimation algorithm for parameters of multivariate Farlie–Gumbel–Morgenstern copula

Abstract This paper focuses on the parameter estimation for d -variate Farlie–Gumbel–Morgenstern (FGM) copula ( $$d\ge 2$$ d ≥ 2 ), which has $$2^d-d-1$$ - 1 dependence parameters to be estimated; therefore, maximum likelihood is not practical a large from viewpoint of computational complexity. Besides, restriction FGM copula’s becomes increasingly complex as large, makes difficult. We propose an effective algorithm by using method inference functions margins under parameters. then discuss its asymptotic normality well performance determined through simulation studies. The proposed also applied real data analysis bearing reliability.