Abstract
Given a function f and weights w on the vertices of a directed acyclic graph G, an isotonic regression of (f,w) is an order-preserving real-valued function that minimizes the weighted distance to f among all order-preserving functions. When the distance is given via the supremum norm there may be many isotonic regressions. One of special interest is the strict isotonic regression, which is the limit of p-norm isotonic regression as p approaches infinity. Algorithms for determining it are given. We also examine previous isotonic regression algorithms in terms of their behavior as mappings from weighted functions over G to isotonic functions over G, showing that the fastest algorithms are not monotonic mappings. In contrast, the strict isotonic regression is monotonic.
Similar content being viewed by others
References
Ayer, M., Brunk, H.D., Ewing, G.M., Reid, W.T., Silverman, E.: An empirical distribution function for sampling with incomplete information. Ann. Math. Stat. 5, 641–647 (1955)
Barlow, R.E., Bartholomew, D.J., Bremner, J.M., Brunk, H.D.: Statistical Inference Under Order Restrictions: The Theory and Application of Isotonic Regression. Wiley, New York (1972)
Gebhardt, F.: An algorithm for monotone regression with one or more independent variables. Biometrika 57, 263–271 (1970)
Hanson, D.L., Pledger, G., Wright, F.T.: On consistency in monotonic regression. Ann. Stat. 1, 401–421 (1973)
Robertson, T., Wright, F.T., Dykstra, R.L.: Order Restricted Statistical Inference. Wiley, New York (1988)
Ubhaya, V.A.: Isotone optimization, I. J. Approx. Theory 12, 146–159 (1974)
Ubhaya, V.A.: Isotone optimization, II. J. Approx. Theory 12, 315–331 (1974)
Punera, K., Ghosh, J.: Enhanced hierarchical classification via isotonic smoothing. In: Proc. Int’l. Conf. World Wide Web, pp. 151–160 (2008)
Zheng, Z., Zha, H., Sun, G.: Query-level learning to rank using isotonic regression. In: Proc. 46th Allerton Conf., pp. 1108–1115 (2008)
Angelov, S., Harb, B., Kannan, S., Wang, L.-S.: Weighted isotonic regression under the L 1 norm. Symp. Discrete Algorithms, 783–791 (2006)
Barlow, R.E., Ubhaya, V.A.: Isotonic approximation. In: Rustagi, J.S. (ed.) Proc. Optimizing Methods in Stat., pp. 77–86. Academic Press, San Diego (1971)
Stout, Q.F.: Weighted L∞ isotonic regression, submitted (2011)
Kaufman, Y., Tamir, A.: Locating service centers with precedence constraints. Discrete Appl. Math. 47, 251–261 (1993)
Cuesta, J.A., Matrán, C.: Conditional bounds and best L ∞-approximations in probability spaces. J. Approx. Theory 56, 1–12 (1989)
Darst, R.B., Sahab, S.: Approximations of continuous and quasi-continuous functions by monotone functions. J. Approx. Theory 38, 9–27 (1983)
Legg, D., Townsend, D.: Best monotone approximation in L ∞[0,1]. J. Approx. Theory 42, 30–35 (1984)
Pólya, G.: Sur une algorithme toujours convergent pour obtenir les polynomes de meilleure approximation de Tchebycheff pour un fonction continue quelconque. Comptes Rendus 157, 840–843 (1913)
Ahuja, R.K., Olin, J.B.: A fast scaling algorithm for minimizing separable convex functions subject ot chain constraints. Oper. Res. 49, 784–789 (2001)
Stout, Q.F.: Unimodal regression via prefix isotonic regression. Comput. Stat. Data Anal. 53, 289–297 (2008)
Thompson, W.A. Jr.: The problem of negative estimates of variance components. Ann. Math. Stat. 33, 273–289 (1962)
Pardalos, P.M., Xue, G.: Algorithms for a class of isotonic regression problems. Algorithmica 23, 211–222 (1999)
Megido, N.: Linear-time algorithms for linear programming in R 3 and related problems. SIAM J. Comput. 12, 759–776 (1983)
Stout, Q.F.: Isotonic regression for multiple independent variables, submitted (2011)
Jackson, D.: Note on the median of a set of numbers. Bull. Am. Math. Soc. 27, 160–164 (1921)
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Panos M. Pardalos.
Rights and permissions
About this article
Cite this article
Stout, Q.F. Strict L∞ Isotonic Regression. J Optim Theory Appl 152, 121–135 (2012). https://doi.org/10.1007/s10957-011-9865-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10957-011-9865-8