skfolio.utils.stats.n_bins_knuth

skfolio.utils.stats.n_bins_knuth(x)

Compute the optimal histogram bin size using Knuth’s rule [1].

The number of bins is bounded to \([1, n]\), where \(n\) is the number of observations. Data with many repeated values, such as zero-filled returns, push the estimate toward the upper bound.

Parameters:

xndarray of shape (n_observations,)

The input array.

Returns:

n_binsint

The optimal bin size.

References

[1]

“Optimal Data-Based Binning for Histograms”. Knuth.