【摘要】 本文中对于多信息源的信息差异性度量B,我们研究了B(u1,u2,…,us)的渐近分布,其中u1,u2,…us是某个概率分布的样本估计．在某些较弱的条件下,我们证明了nB(u1,u2,…,us)的极限分布是自由度为1的X2分布的加权和,其中n是样本容量,u1,u2,…,us不必相互独立．更多还原

【Abstract】 For a measure of information discrepancy B, in this paper we investigate the asymptotic distributions of B(u1 ,u2,…, us), where u1,u2,…, us are sample-based estimators of a common probability distribution. Under some mild conditions, we prove that the limiting distribution of nB(u1,u2,…, us) is the one of weighted sum of chi square variables with one degree of freedom, where u1,u2,…, us may not be independent of one another and n is the sample size.更多还原