Compute the number of classes for a histogram, the number of nodes of a binning grid, etc.
rule(x, d = 1, rule = c("Rice", "Sturges", "scott", "FD"), ...) rule.binning(x, ...) # S3 method for default rule.binning(x, d = ncol(x), a = 2, b = d + 1, ...) rule.svar(x, ...) # S3 method for default rule.svar(x, d = ncol(x), a = 2, b = d + 1, ...) # S3 method for bin.den rule.svar(x, ...)
| x | data vector or object used to select a method. |
|---|---|
| d | (spatial) dimension. |
| rule | character; rule to be used. |
| ... | further arguments passed to or from other methods. |
| a | scale values. |
| b | exponent values. |
The rule values (vector or scalar).
rule.binning returns a vector with the suggested number of bins
on each dimension.
rule.binning.default returns rep(ceiling(a * nrow(x) ^ (1 / b)), d).
rule.svar returns the suggested number of bins
for variogram estimation.
rule.svar.default returns ceiling(a * (nrow(x)^2 / 4) ^ (1 / b)).
The Rice Rule, \(m = \lceil 2 n^{1/3} \rceil,\)
is a simple alternative to Sturges's rule (nclass.Sturges).