STK_FILLDIST computes the fill distance of a set of points

 CALL: FD = stk_filldist(X, BOX)

    computes the fill distance FD of the dataset X in the hyper-rectangle
    BOX, using the computational-geometric algorithm of L. Pronzato and
    W. G. Muller [1]. Recall that

       D = max_{Y in BOX} min_{1 <= i <= n} norm(X(i,:) - Y),         (1)

    where norm(.) denotes the Euclidean norm in R^d. Optimal designs with
    respect to the fill distance are sometimes called "minimax" designs
    (see, e.g., [2]).

 CALL: FD = stk_filldist(X)

    assumes that the fill distance is to be computed with respect to the
    hyperrectangle BOX = [0; 1]^d.

 CALL: FD = stk_filldist(X, Y)

    computes the fill distance FD of X using the "test set" Y. More preci-
    sely, if X and Y are respectively n x d and m x d, then

       FD = max_{1 <= j <= m} min_{1 <= i <= n} norm(X(i,:) - Y(j,:)),

    If Y is dense enough in some subset BOX of R^d, then FD should be close
    to the actual fill distance of X in BOX.

 CALL: [FD, YMAX] = stk_filldist(...)

    also returns the point YMAX where the maximal distance is attained.

 NOTE:

    stk_filldist is actually a wrapper around stk_filldist_discretized and
    stk_filldist_exact. Which function to call is guessed based on the number
    of rows of the second argument. Because of that, the test set Y is required
    to have at least 3 rows.

 REFERENCES

   [1] Luc Pronzato and Werner G. Muller, "Design of computer
       experiments: space filling and beyond", Statistics and Computing,
       22(3):681-701, 2012.

   [2] Mark E. Johnson, Leslie M. Moore and Donald Ylvisaker, "Minimax
       and maximin distance designs", Journal of Statistical Planning
       and Inference, 26(2):131-148, 1990.

 See also: stk_dist, stk_mindist, stk_filldist_exact, stk_filldist_discretized

Package: stk