Defined as
$$R_{\textnormal{high}} (p,k)_n := \frac{\sum_{i=2k}^n \left| \Delta_{i,k}^{n,2} X \right|^p}
{\sum_{i=k}^n \left| \Delta_{i,k}^{n,1} X \right|^p}, \qquad$$
$$R_{\textnormal{low}} (p,k)_n := \frac{\sum_{i=2k}^n \left| \Delta_{i,k}^{2} X \right|^p}
{\sum_{i=k}^n \left| \Delta_{i,k}^{1} X \right|^p}$$
Usage
R_hl(p, k, path)
Arguments
p
power
k
increment order
path
sample path of lfsm on which the inference is to be performed
Details
The computation procedure for high- and low frequency cases is the same, since there is no way to control frequency given a sample path.