In this paper, the solution of the asymmetric eigenvalue complementarity problem (EiCP) is investigated by means of a variational inequality formulation. This problem is then solved by finding a stationary point of the gap function and the regularized gap function. A nonlinear programming formulation of the EiCP results from the gap function. A hybrid algorithm combining a projection technique and a modified Josephy-Newton method is proposed to solve the EiCP by finding a stationary point of the regularized gap function. Numerical results show that the proposed method can in general solve EiCPs efficiently.