Entropy and optimal partition for data analysis

The concept of symbolic dynamics, entropy and complexity measures has been widely utilized for the analysis of measured time series. However, little a ttention as been devoted to investigate the effects of choosing different p artitions to obtain the coarse-grained symbolic sequences. Because the theo retical concepts of generating partitions mostly fail in the case of empiri cal data, one commonly introduces a homogeneous partition which ensures rou ghly equidistributed symbols. Wie will show that such a choice may lead to spurious results for the estimated entropy and will not fully reveal the ra ndomness of the sequence.