The Annals of Statistics

On multivariate quantiles under partial orders

Alexandre Belloni and Robert L. Winkler

Full-text: Open access

Abstract

This paper focuses on generalizing quantiles from the ordering point of view. We propose the concept of partial quantiles, which are based on a given partial order. We establish that partial quantiles are equivariant under order-preserving transformations of the data, robust to outliers, characterize the probability distribution if the partial order is sufficiently rich, generalize the concept of efficient frontier, and can measure dispersion from the partial order perspective.

We also study several statistical aspects of partial quantiles. We provide estimators, associated rates of convergence, and asymptotic distributions that hold uniformly over a continuum of quantile indices. Furthermore, we provide procedures that can restore monotonicity properties that might have been disturbed by estimation error, establish computational complexity bounds, and point out a concentration of measure phenomenon (the latter under independence and the componentwise natural order).

Finally, we illustrate the concepts by discussing several theoretical examples and simulations. Empirical applications to compare intake nutrients within diets, to evaluate the performance of investment funds, and to study the impact of policies on tobacco awareness are also presented to illustrate the concepts and their use.

Article information

Source
Ann. Statist., Volume 39, Number 2 (2011), 1125-1179.

Dates
First available in Project Euclid: 9 May 2011

Permanent link to this document
https://projecteuclid.org/euclid.aos/1304947046

Digital Object Identifier
doi:10.1214/10-AOS863

Mathematical Reviews number (MathSciNet)
MR2816350

Zentralblatt MATH identifier
1216.62082

Subjects
Primary: 62H12: Estimation 62J99: None of the above, but in this section
Secondary: 62J07: Ridge regression; shrinkage estimators

Keywords
Multivariate quantiles partial order uniform estimation

Citation

Belloni, Alexandre; Winkler, Robert L. On multivariate quantiles under partial orders. Ann. Statist. 39 (2011), no. 2, 1125--1179. doi:10.1214/10-AOS863. https://projecteuclid.org/euclid.aos/1304947046


Export citation

References

  • [1] Alexander, K. S. (1987). Rates of growth and sample moduli for weighted empirical processes indexed by sets. Probab. Theory Related Fields 75 379–423.
  • [2] Bagchi, S. C. and Sitaram, A. (1982). Determining sets for measures on ℝn. Illinois J. Math. 26 419–422.
  • [3] Barnett, V. (1976). The ordering of multivariate data. J. Roy. Statist. Soc. Ser. A 139 318–355.
  • [4] Blum, L., Cucker, F., Shub, M. and Smale, S. (1998). Complexity and Real Computation. Springer, New York.
  • [5] Bochner, S. (1937). Bounded analytic functions in several variables and multiple Laplace integrals. Amer. J. Math. 59 732–738.
  • [6] Buchinsky, M. and Hahn, J. (1998). An alternative estimator for the censored quantile regression model. Econometrica 66 653–671.
  • [7] Chakraborty, B. (2003). On multivariate quantile regression. J. Statist. Plann. Inference 110 109–132.
  • [8] Chaudhuri, P. (1996). On a geometric notion of quantiles for multivariate data. J. Amer. Statist. Assoc. 91 862–872.
  • [9] Cheng, Y. and De Gooijer, J. G. (2007). On the uth geometric conditional quantile. J. Statist. Plann. Inference 137 1914–1930.
  • [10] Chernozhukov, V., Fernández-Val, I. and Galichon, A. (2009). Improving point and interval estimators of monotone functions by rearrangement. Biometrika 96 559–575.
  • [11] Chernozhukov, V., Fernández-Val, I. and Galichon, A. (2010). Quantile and probability curves without crossing. Econometrica 78 1093–1125.
  • [12] Chernozhukov, V., Hong, H. and Tamer, E. (2007). Estimation and confidence regions for parameter sets in econometric models. Econometrica 75 1243–1284.
  • [13] de Andrade Jr., F. (2011). Measures of downside risk and mutual fund flows. Kellogg School of Management, Northwestern Univ. Unpublished manuscript.
  • [14] Dette, H. and Volgushev, S. (2008). Non-crossing non-parametric estimates of quantile curves. J. R. Stat. Soc. Ser. B Stat. Methodol. 70 609–627.
  • [15] Di Bucchianico, A., Einmahl, J. H. J. and Mushkudiani, N. A. (2001). Smallest nonparametric tolerance regions. Ann. Statist. 29 1320–1343.
  • [16] Dudley, R. M. (1999). Uniform Central Limit Theorems. Cambridge Studies in Advanced Mathematics 63. Cambridge Univ. Press, Cambridge.
  • [17] Dudley, R. M. (2002). Real Analysis and Probability. Cambridge Studies in Advanced Mathematics 74. Cambridge Univ. Press, Cambridge.
  • [18] Einmahl, J. H. J. and Mason, D. M. (1992). Generalized quantile processes. Ann. Statist. 20 1062–1078.
  • [19] Flay, B., Brannon, B., Johnson, C., Hansen, W., Ulene, A., Whitney-Saltiel, D., Gleason, L., Sussman, S., Gavin, M., Kimarie, G., Sobol, D. and Spiegel, D. (1988). The television school and family smoking prevention and cessation project: I. Theoretical basis and program development. Preventive Medicine 17 585–607.
  • [20] Gardner, R. J. (2002). The Brunn–Minkowski inequality. Bull. Amer. Math. Soc. (N.S.) 39 355–405.
  • [21] Gibbons, R. D. and Hedeker, D. (1997). Random effects probit and logistic regression models for three-level data. Biometrics 53 1527–1537.
  • [22] Giné, E. and Koltchinskii, V. (2006). Concentration inequalities and asymptotic results for ratio type empirical processes. Ann. Probab. 34 1143–1216.
  • [23] Grötschel, M., Lovász, L. and Schrijver, A. (1984). Geometric methods in combinatorial optimization. In Progress in Combinatorial Optimization (Waterloo, Ont., 1982) 167–183. Academic Press, Toronto, ON.
  • [24] Hallin, M., Paindaveine, D. and Šiman, M. (2010). Multivariate quantiles and multiple-output regression quantiles: From L1 optimization to halfspace depth. Ann. Statist. 38 635–669.
  • [25] Hallin, M., Paindaveine, D. and Siman, M. (2010). Rejoinder. Ann. Statist. 38 694–703.
  • [26] Henriksson, R. and Merton, R. (1981). On market timing and investment performance. II. Statistical procedures for evaluating forecasting skills. J. Bus. 54 513–533.
  • [27] Huber, P. J. (1973). Robust regression: Asymptotics, conjectures and Monte Carlo. Ann. Statist. 1 799–821.
  • [28] Jagannathan, R. and Korajczyk, R. (1986). Assesing the market timing performance of managed portfolios. J. Bus. 59 217–235.
  • [29] Kalai, A. T. and Vempala, S. (2006). Simulated annealing for convex optimization. Math. Oper. Res. 31 253–266.
  • [30] Keeney, R. L. and Raiffa, H. (1976). Decisions with Multiple Objectives: Preferences and Value Tradeoffs. Wiley, New York.
  • [31] Kim, J. and Pollard, D. (1990). Cube root asymptotics. Ann. Statist. 18 191–219.
  • [32] Koenker, R. (2005). Quantile Regression. Econometric Society Monographs 38. Cambridge Univ. Press, Cambridge.
  • [33] Koenker, R. and Bassett, G. Jr. (1978). Regression quantiles. Econometrica 46 33–50.
  • [34] Kong, L. and Mizera, I. (2010). Discussion of “Multivariate quantiles and multiple-output regression quantiles: From L1 optimization to halfspace depth,” by M. Hallin, D. Paindaveine and M. Šiman. Ann. Statist. 38 685–693.
  • [35] Lorentz, G. G. (1953). An inequality for rearrangements. Amer. Math. Monthly 60 176–179.
  • [36] Lovász, L. and Vempala, S. (2006). Fast algorithms for logconcave functions: Sampling, rounding, integration and optimization. In Proceedings of the 47th Annual IEEE Symposium on Foundations of Computer Science (FOCS) 57–68. IEEE Computer Society, Los Alamitos, CA.
  • [37] Molchanov, I. S. (2005). Theory of Random Sets. Springer, London.
  • [38] Motwani, R. and Raghavan, P. (1995). Randomized Algorithms. Cambridge Univ. Press, Cambridge.
  • [39] Neocleous, T. and Portnoy, S. (2008). On monotonicity of regression quantile functions. Statist. Probab. Lett. 78 1226–1229.
  • [40] Neocleous, T., Van den Branden, K. and Portnoy, S. (2006). Correction to “Censored regression quantiles,” by S. Portnoy. J. Amer. Statist. Assoc. 101 860–861.
  • [41] Newey, W. K. and Powell, J. L. (1990). Efficient estimation of linear and type I censored regression models under conditional quantile restrictions. Econometric Theory 6 295–317.
  • [42] Nusser, S. M., Carriquiry, A. L., Dodd, K. W. and Fuller, W. A. (1996). A semiparametric transformation approach to estimating usual daily intake distributions. J. Amer. Statist. Assoc. 91 1440–1449.
  • [43] Pollard, D. (1995). Uniform ratio limit theorems for empirical processes. Scand. J. Statist. 22 271–278.
  • [44] Polonik, W. and Yao, Q. (2002). Set-indexed conditional empirical and quantile processes based on dependent data. J. Multivariate Anal. 80 234–255.
  • [45] Portnoy, S. (2003). Censored regression quantiles. J. Amer. Statist. Assoc. 98 1001–1012.
  • [46] Powell, J. L. (1984). Least absolute deviations estimation for the censored regression model. J. Econometrics 25 303–325.
  • [47] Powell, J. L. (1986). Censored regression quantiles. J. Econometrics 32 143–155.
  • [48] Powell, J. L. (1986). Symmetrically trimmed least squares estimation for Tobit models. Econometrica 54 1435–1460.
  • [49] Powell, J. L. (1991). Estimation of monotonic regression models under quantile restrictions. In Nonparametric and Semiparametric Methods in Econometrics and Statistics (Durham, NC, 1988) 357–384. Cambridge Univ. Press, Cambridge.
  • [50] Serfling, R. (2002). Quantile functions for multivariate analysis: Approaches and applications. Statist. Neerlandica 56 214–232.
  • [51] Serfling, R. (2011). Equivariance and invariance properties of multivariate quantile and related functions, and the role of standardization. Dept. Mathematical Sciences, Univ. Texas at Dallas and Michigan State Univ. Unpublished manuscript.
  • [52] Serfling, R. and Zuo, Y. (2010). Discussion of “Multivariate quantiles and multiple-output regression quantiles: From L1 optimization to halfspace depth,” by M. Hallin, D. Paindaveine and M. Šiman. Ann. Statist. 38 676–684.
  • [53] Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium under conditions of risk. Journal of Finance 19 425–442.
  • [54] Sitaram, A. (1983). A theorem of Cramér and Wold revisited. Proc. Amer. Math. Soc. 87 714–716.
  • [55] Sitaram, A. (1984). Fourier analysis and determining sets for Randon measures on ℝn. Illinois J. Math. 28 339–347.
  • [56] U.S. Department of Agriculture, Human Nutrition Information Service. (1987). Continuing Survey of Food Intakes by Individuals, Women 19–50 Years and Their Children 1–5 Years, 4 Days, 1985. CSFII Report No. 85-4. U.S. Government Printing Office, Washington, DC.
  • [57] van der Vaart, A. W. and Wellner, J. A. (1996). Weak Convergence and Empirical Processes: With Applications to Statistics. Springer, New York.
  • [58] Vempala, S. (2005). Geometric random walks: A survey. In Combinatorial and Computational Geometry. Mathematical Sciences Research Institute Publications 52 577–616. Cambridge Univ. Press, Cambridge.
  • [59] Wei, Y. (2010). Discussion of “Multivariate quantiles and multiple-output regression quantiles: From L1 optimization to halfspace depth,” by M. Hallin, D. Paindaveine and M. Šiman. Ann. Statist. 38 670–675.
  • [60] Wermers, R. (2000). Mutual fund performance: An empirical decomposition into stock-picking talent, style, transaction costs, and expenses. Journal of Finance 55 1655–1695.
  • [61] Womersley, R. S. (1986). Censored discrete linear l1 approximation. SIAM J. Sci. Statist. Comput. 7 105–122.