Almost Sure Limit Behavior of Discrete Random Sequence with Application to Arbitrary Information Sources

作者：Wang Zhong Zhi^*; Ding Keyue

来源：Applied Mathematics and Information Sciences, 2008, 2(3): 333-343.

• Random sequence
• random limit logarithmic likelihood ratio
• strong deviation theorem
• conditional entropy

In this paper, the notion of the limit of logarithmic likelihood ratio of random sequences, as a measure of dissimilarity between two probability measures, is introduced. After establishing a ratio of two measures by means of constructing a new probability measure, we obtained the strong random deviation theorems for partial sums of functions of arbitrary discrete random variables under suitable restrictive conditions. As a direct application, we used our results to derive some limit properties of discrete information source.

• 出版日期2008-9
• 单位安徽工业大学