Keeping the data set size constant approximately equalises the entropy of the data given the generating model as illustrated in Additional file 1: Figure S9.
Although it is possible to manually chose an appropriate discretization for the data given the size of the network, we followed a systematic procedure to find the most appropriate discretization.
Therefore, the model giving the smallest value of the code is to be considered the best model explaining the data given the sample size.
The data give particle number concentrations (particles per cubic centimeter of air) versus particle diameter across the size distribution.
As a result, ApEn can be sensitive to the size of the data set, giving inappropriately low values when the total number of data points is low; this, and a lack of consistency in differentiating signals when m and r are altered, have led to the development of a new family of statistics named sample entropy (SampEn), in which self-matches are excluded in the analysis [ 110].
That is, when a data object is updated at the server, the server sends an invalidation report to the MH through the proxy to invalidate the cached data object, provided that the size of the data object exceeds the given threshold.
For living cells the potential significance of the photoselection effect in interpretation of GP data, given the large size and irregular geometry of cells and especially when considering phenomena using 3 or 4-dimensional microscopy deserves careful consideration.
We also observed a cluster of mitochondrial proteins including NDUFA11 near COX6A2 and COXA5, although given the size of the data this may have been expected by chance alone.
In a typical pattern classification scenario the available data set D is finite and is defined as D = { ( x i, y i ) } i = 1 N, where x i ∈X; y i ∈Y and N refers to the size of the given sample data.
Given the size of the data base, 99.9% limits were chosen to yield conservative results.
