Examining Human Unipedal Quiet Stance: Characterizing Control through Jerk

Figure 7

Typical characteristics of data tested using Algorithm 1 with the null hypothesis failing to be rejected. (a) This is an original A-P time series that was tested and (b) its corresponding power spectral density. (c) Quantile-quantile plot of the original data: this indicates that these data follow a fairly normal distribution with a tendency toward fat tails. (d) Again, as with Algorithm 0, a rank-ordered criterion is applied and, here, a failure to reject the null hypothesis is indicated. Sample entropy (with a tolerance 0.20 of the standard deviation of the data values) was used. This corresponds to the original time series having a character consistent with linearly filtered IID noise. (e, f) The power spectral density does show an interesting falloff. The probability distributions for the original and, respectively, one of the surrogates are displayed. They are different as they are expected to use Algorithm 1. However, the difference, at least visually, is not too drastic (as mentioned in the text, this has to do with worries of falsely rejecting the null hypothesis under Algorithm 1, although, in this particular case, it is not a concern [24]).

(a)

(b)

(c)

(d)

(e)

(f)