One of the most important bivariate distributions in statistical inference is the bivariate normal distribution. In this paper we consider ML estimation of parameters of bivariate normal distribution using bivariate record values. The asymptotic properties especially asymptotic efficiencies of the ML estimators are studied and calculated by deriving the Fisher information matrix about the five dimension parameter vector contained in the vector of first n bivariate record values. The relative asymptotic efficiencies are concerned with respect to ML estimators based on a bivariate sample of size n and the corresponding ML estimators based on bivariate record values and inter-record times.

Keywords