We consider the problem of estimating two isotonic regression curves g1*
and g2* under the constraint that they are ordered, i.e. g1* <= g2*. Given
two sets of n data points y_1, ..., y_n and z_1, ..., z_n that are
observed at (the same) deterministic design points x_1, ..., x_n, the
estimates are obtained by minimizing the Least Squares criterion L(a, b) =
sum_{i=1}^n (y_i - a_i)^2 w1(x_i) + sum_{i=1}^n (z_i - b_i)^2 w2(x_i) over
the class of pairs of vectors (a, b) such that a and b are isotonic and
a_i <= b_i for all i = 1, ..., n.
We