Isotonic Regression

In numerical analysis, isotonic regression (IR) involves finding a weighted least-squares fit to a vector with weights vector subject to a set of non-contradictory constraints of kind .

Such constraints define partial order or total order and can be represented as a directed graph, where N is the set of variables involved, and E is the set of pairs (i, j) for each constraint . Thus, the IR problem corresponds to the following quadratic program (QP):

In the case when is a total order, a simple iterative algorithm for solving this QP is called the pool adjacent violators algorithm (PAVA). Best and Chakravarti (1990) have studied the problem as an active set identification problem, and have proposed a primal algorithm in O(n), the same complexity as the PAVA, which can be seen as a dual algorithm.

IR has applications in statistical inference, for example, to fit of an isotonic curve to mean experimental results when an order is expected.

Another application is nonmetric multidimensional scaling (Kruskal, 1964), where a low-dimensional embedding for data points is sought such that order of distances between points in the embedding matches order of dissimilarity between points. Isotonic regression is used iteratively to fit ideal distances to preserve relative dissimilarity order.

Isotonic regression is also sometimes referred to as monotonic regression. Correctly speaking, isotonic is used when the direction of the trend is strictly increasing, while monotonic could imply a trend that is either strictly increasing or strictly decreasing.

Isotonic Regression under the for is defined as follows: