Abstract
For F1, F2 being distribution functions such that F2 is absolutely continuous w.r.t. F1 and F1 is continuous, the chi-square divergence between F2 and F1 equal to is considered. Its estimator
based on the modified kernel estimate of the grade density g is introduced. Asymptotic normality of
is established, asymptotic variance turns out to be larger than for the estimator of χ2 in the case when the i.i.d. sample from g is observable. Finally, choice of smoothing parameter is discussed based on calculation of the asymptotic mean square error for some approximation of χ2.