Abstract
Consider the nonparametric regression modelY=g 0(T)+u, whereY is real-valued,u is a random error,T ranges over a nondegenerate compact interval, say [0,1], andg 0(·) is an unknown regression function, which ism(m≥0) times continuously differentiable and itsmth derivative,g (m)0 , satisfies a Hölder condition of order γ(m+γ>1/2). A piecewise polynomialL 1-norm estimator ofg 0 is proposed. Under some regularity conditions including that the random errors are independent but not necessarily have a common distribution, it is proved that the rates of convergence of the piecewise polynomialL 1-norm estimator are\(o(n^{ - \frac{{m + \gamma - 1/2 - \delta }}{{2\left( {m + \gamma } \right) + 1}}} )\) almost surely and\(o(n^{ - \frac{{m + \gamma - \delta }}{{2\left( {m + \gamma } \right) + 1}}} )\) in probability, which can arbitrarily approach the optimal rates of convergence for nonparametric regression, where σ is any number in (0, min((m+γ−1/2)/3,γ)).
Similar content being viewed by others
References
Stone, C.,Additive regression and other nonparametric models, Ann. Statist.,13 (1985), 689–705.
Härdle, W., Applied Nonparametric Regression, Cambridge University Press, 1990.
Friedman, J. H.,Multivariate additive regression splines, (with discussion), Ann. Statist.,19 (1991), 1–141.
Major, P.,On a nonparametric estimation of the regression function, Studia Sci. Math. Hungar.,8 (1973), 347–361.
Stone, C.,Optimal rates of convergence for nonparametric estimators, Ann. Statist.,8 (1980), 1348–1360.
Stone, C.,Optimal rates of convergence for nonparametric estimators, Ann. Statist.,10 (1982), 1040–1053.
Chen, K.W.,Asymptotic optimal selection of a piecewise polynomial estimator of a regression function, J. Multivariate Anal.,22 (1987), 230–244.
McGee, V.E. and Carleton, W.T.,Piecewise regression, J. Amer. Assoc.,65 (1970) 1109–1124.
Lerman, P.M.,Fitting segmented regression model by grid search, Appl. Statist.,29 (1980), 77–84.
Tisher, A. and Zang, I.,A maximum likelihood method for piecewise regression models with a continuous dependent variable, Appl. Statist.,30 (1981), 116–124.
Koenker, R. and D'Orey, V.,Computing regression quantiles, Appl. Statist.,36 (1987), 383–393.
Gentle, J.E., Narula, S. C., and Sposito, V. A., Algorithms for unconstrainedL 1 linear regression, in Statistical Data Analysis Based onL 1-norm and Related Methods (Y. Dodge eds), 1987.
Oberhofer, W.,The consistency of nonlinear regression minimizing the L 1-norm, Ann. Statist.,10 (1982), 316–319.
Wu, Y. H.,Strong consistency and exponential rate of the “minimum L 1-norm” estimates in linear regression models, Comput. Statist. & Data Anal.,6 (1988), 285–295.
Chen, X. R., Bai, Z.D., Zhao, L. C., and Wu, Y. H.,The asymptotic normality of L 1-norm estimators in linear models, Science in China, A No.5 (1990), 449–462.
Shi, P.D. and Li, G.Y., Optimal global rates of convergence ofM-estimates for nonparametric regression, to appear in J. Syc & Math. Scis., 1991.
Pollard, D., Convergence of Stochastic Processes, Springer-Verlag, New York, 1984.
Author information
Authors and Affiliations
Additional information
Supported by the National Natural Science Foundation of China.
Rights and permissions
About this article
Cite this article
Peide, S., Ping, C. Asymptotics of the “minimumL 1-norm” estimates in nonparametric regression models. Acta Mathematica Sinica 10, 276–288 (1994). https://doi.org/10.1007/BF02560718
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02560718