Table 13.

Summary of the conclusions of the paper and the supporting information.

	Recommendation	Supporting Data	Justification
PE (absolute value)	$N>>m!$	Figure 3, Figure 4, Figure 5, Figure 6 and Figure 7	Pattern probability estimation in other works.
PE (relative value) For classification	$N=200$	Figure 8, Figure 9 and Figure 10 Very similar results in other studies ([24,26,27]). Table 2, Table 3, Table 4, Table 5, Table 6 and Table 7, Table 9, Table 10, Table 11 and Table 12. Very similar results for 10 datasets exhibiting a varied and diverse set of features and properties.	Class differences are present at any length in stationary records. Long records are usually non-stationary. There are forbidden patterns. No need to look for them. Not all the ordinal patterns are representative of the differences. Real signals are mostly chaotic.