'mean(X)' is the mean value for each element X, 'sd(X)' the
standard deviation, and 'X[n]' the nth or latest observed values.
'res' is the residual X[n] - mean(X) and represents a centred measure of
deviation for the latest observed value.
The 'z-score' (or standard score) is calculated as res / sd(X) and is a
centred and scaled measure of deviation for the latest observed value.
'p >= |z|' represents the empirical probability of the latest observed
absolute 'z-score' or greater.
'p*' will display a star if 'p >= |z|' is less than or equal to the value
of the argument 'signif'.
'E(|res|)|p' represents the mean or expected absolute value of 'res',
conditional upon the absolute 'z-score' being greater than equal to the
latest observed absolute 'z-score'.