generates random variates with multivariate Lipschitz density

 RanLip generates random variates with an arbitrary multivariate
 Lipschitz density.
 .
 While generation of random numbers from a variety of distributions is
 implemented in many packages (like GSL library
 http://www.gnu.org/software/gsl/ and UNURAN library
 http://statistik.wu-wien.ac.at/unuran/), generation of random variate
 with an arbitrary distribution, especially in the multivariate case, is
 a very challenging task. RanLip is a method of generation of random
 variates with arbitrary Lipschitz-continuous densities, which works in
 the univariate and multivariate cases, if the dimension is not very
 large (say 3-10 variables).
 .
 Lipschitz condition implies that the rate of change of the function (in
 this case, probability density p(x)) is bounded:
 .
 |p(x)-p(y)|<M||x-y||.
 .
 From this condition, we can build an overestimate of the density, so
 called hat function h(x)>=p(x), using a number of values of p(x) at some
 points. The more values we use, the better is the hat function. The
 method of acceptance/rejection then works as follows: generatea random
 variate X with density h(x); generate an independent uniform on (0,1)
 random number Z; if p(X)<=Z h(X), then return X, otherwise repeat all
 the above steps.
 .
 RanLip constructs a piecewise constant hat function of the required
 density p(x) by subdividing the domain of p (an n-dimensional rectangle)
 into many smaller rectangles, and computes the upper bound on p(x)
 within each of these rectangles, and uses this upper bound as the value
 of the hat function.