S-SMART最大熵法及在小样本中的应用

Application of S-SMART Maximum Entropy Method in Small Sample

基于S-SMART方法在解决小样本问题中的优势,结合信息论中熵的概念,找出了一种S-SMART最大熵法来确定Bayes先验分布的途径。首先将样本扩大,然后利用最大熵法来确定先验分布。该方法能较好地避免主观因素的影响,对于解决小样本问题非常有效。模拟不同的放大倍数均得到了较好的结果,证明了该方法的可行性与有效性。

Based on the S-SMART methods to solve the problem of small sample advantage, and in the concept of entropy based on information, the S-SMART maximum entropy method is used to deter- mine the Bayes prior distribution. First it will expand the sample＇ s volume, and then use the maxi- mum entropy method to determine the prior distribution. This method can avoid the subjective factors and is effective in solving the problem of small sample. Through the simulation of different amplifica- tion got good results show the feasibility and validity of this method.