Abstract
This paper considers a nonparametric varying coefficient regression with spatial data. A global smoothing procedure is developed by using B-spline function approximations for estimating the coefficient functions. Under mild regularity assumptions, the global convergence rates of the B-spline estimators of the unknown coefficient functions are established. Asymptotic results show that our B-spline estimators achieve the optimal convergence rate. The asymptotic distributions of the B-spline estimators of the unknown coefficient functions are also derived. A procedure for selecting smoothing parameters is given. Finite sample properties of our procedures are studied through Monte Carlo simulations. Application of the proposed method is demonstrated by examining voting behaviors across US counties in the 1980 presidential election.
Similar content being viewed by others
References
Cressie N A C. Statistics for Spatial Data. New York: Wiley, 1991
Basawa I V. Special issue on spatial statistics. Part 1. J Statist Plann Infer, 50: 311–411 (1996)
Basawa I V. Special issue on spatial statistics. Part 2. J Statist Plann Infer, 51: 1–97 (1996)
Guyon X. Random Fields on a Network. New York: Springer, 1995
Tran L T. Kernel density estimation on random field. J Multivariate Anal, 34: 37–53 (1990)
Fan J, Zhang W. Statistical estimation in varying coefficient models. Ann Statist, 27: 1491–1518 (1999)
Huang J Z, Wu C O, Zhou L. Polynomial spline estimation and inference for varying coefficient models with longitudinal data. Statist Sin, 14: 763–788 (2004)
Tang Q, Wang J. One-step estimation for varying coefficient models. Sci China Ser A, 48(2): 198–213 (2005)
Tang Q, Wang J. L 1-estimation for varying coefficient models. Statistics, 39(5): 389–404 (2005)
Tang Q, Wang J. Reducing component estimation for varying coefficient models with longitudinal data. Sci China Ser A, 51(2): 250–272 (2008)
Tran L T, Yakowitz S. Nearest neighbor estimators for random fields. J Multivariate Anal, 44: 23–46 (1993)
Hallin M, Lu Z, Tran L T. Density estimation for spatial linear processes. Bernoulli, 7: 657–668 (2001)
Hallin M, Lu Z, Tran L T. Local linear spatial regression. Ann Statist, 32: 2469–2500 (2004)
Biau G. Spatial kernel density estimation. Math Methods Statist, 12: 371–390 (2003)
Lu Z, Cheng P. Spatial nonparametric regression estimation: Nonisotropic case. Acta Math Appl Sinica (English Ser), 18: 641–656 (2002)
Lu Z, Cheng P. Spatial kernel regression estimation: Weak consistency. Statist Probab Lett, 68: 125–136 (2004)
Gao J, Lu Z, Tjøstheim D. Estimation in semiparametric spatial regression. Ann Statist, 34: 1395–1435 (2006)
Lu Z, Lundervold A, Tjøstheim D, et al. Exploring spatial nonlinearity using additive approximation. Bernoulli, 13: 447–472 (2007)
Schumaker L L. Spline Functions: Basic Theory. New York: Wiley, 1981
Stone C J. Optimal global rates of convergence for nonparametric regression. Ann Statist, 10: 1040–1053 (1982)
Barrow D L, Smith P W. Asymptotic properties of best L 2[0, 1] approximation by spline with varible knots. Quart Appl Math, 36: 293–304 (1978)
Schwarz G. Estimating the dimension of a model. Ann Statist, 6: 461–464 (1978)
Kelley R P, Barry R. Quick computation of regressions with a spatially autoregressive dependent variable. Geographical Anal, 29: 232–247 (1997)
de Boor C. A Practical Guide to Splines. New York: Springer, 1978
Agarwal G G, Studden W J. Asymptotic integrated mean square error using least squares and bias minimizing splines. Ann Statist, 8: 1307–1325 (1980)
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was partially supported by National Natural Science Foundation of China (Grant No. 10671089), China Postdoctoral Science Foundation and Jiangsu Planned Projects for Postdoctoral Research Funds
Rights and permissions
About this article
Cite this article
Tang, Q., Cheng, L. B-spline estimation for varying coefficient regression with spatial data. Sci. China Ser. A-Math. 52, 2321–2340 (2009). https://doi.org/10.1007/s11425-009-0111-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11425-009-0111-x
Keywords
- spatial data
- varying coefficient regression
- B-spline estimators
- convergence rate
- asymptotic distribution