Statistical Inference for a Kind of Nonlinear Regression Model

Abstract:

As we all know, statistical inference of linear models has been a hot topic of statistical and econometric research. However, in many practical problems, the variable of interest and covariates are often nonlinear relationship. The performance of the statistical inference using linear models model can be very poor. In this paper, the statistical inference of a nonlinear regression model under some additional restricted conditions is investigated. The restricted estimator for the unknown parameter is proposed. Under some mild conditions, the asymptotic normality of the proposed estimator is established on the basis of Lagrange multiplier and hence can be used to construct the asymptotic confidence region of the regression parameter.

• Periodical: Advanced Materials Research (Volumes 616-618)
• Pages: 2149-2152
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.616-618.2149
• Online since: December 2012
• Authors: Yu Ying Jiang, Yong Ming Zhang
• Keywords: Asymptotic Confidence Region, Lagrange Multipliers, Nonlinear Regression Mode, Statistical Inference

References

[1] Kim Y Choi H and Oh H S. Smoothly clipped absolute deviation on high dimensions, J.Amer. Statis. Asso., 2008, 133(484):1665-1673.

DOI: 10.1198/016214508000001066

[2] Liu Q and Xue L G. Empirical likelihood confidence regions for the parameter of a linear EV model with missing data, Math. Pract. Theor. 2008,38(24):147-151 .(In Chinese)

[3] Xue L G. Empirical likelihood for linear models with missing responses, J. Multi. Anal. 2009,100:1353-1366.

[4] Choi N H Li W and Zhu J. Variable selection with the strong heredity constraint and its oracle property, J.Amer. Statis. Asso., 2010, 105(489):354-364.

DOI: 10.1198/jasa.2010.tm08281

[5] Liu Q. Estimation of the linear EV model with censored data, J. Statis. Plan. Infer. 2011, 141(7): 2463- 2471.

[6] Fan J Lv J and Qi L. Sparse high-dimensional models in economics, Annu. Rev. Econ. 2011, 3:291–317.

DOI: 10.1146/annurev-economics-061109-080451

[7] Liu Q. Empirical likelihood inference for linear EV models with missing responses problem, Advan. Mater. Rese., 2012, 482-484: 1999-2002.

DOI: 10.4028/www.scientific.net/amr.482-484.1999

[8] Bates D M. Nonlinear regression analysis and its applications, John Wiley & Sons, 1988.

[9] Hsiao C. Consistent estimation for some nonlinear error- in-variable models. J. Econometrics. 1989,41:159-185.

DOI: 10.1016/0304-4076(89)90047-x

[10] Xue L G and Liao J Y. Parameter estimation of a regression model under censored data. Chinese J.Engin. Math. 2005,22(4):712-718.

[11] Stute W. Xue L G and Zhu L X. Empirical likelihood inference in nonlinear errors-in-covariables models with validation data. J. Amer. Statist. Assoc, 2007,102(477): ~332-346.

DOI: 10.1198/016214506000000816

[12] Liu Q. Estimation of the nonlinear EV models with missing response data，J. Fuzhou Univer., Natur. Scien., 2010, 38(4):491-497. (In Chinese)

[13] Wei C H and Wu X Z. Profile lagrange multiplier test for partially linear varying-coefficient regression models. J.Sys.Sci. Math.Scis. 2008,18(4):416-424. (In Chinese)

[14] Zhang W W Li G R and Xue L G. Profile inference on partially linear varying-coefficient errors-in-variables models under restricted condition, Comp. Stat. Data Ana. 2011, DOI: 10.1016 /j.csda.2011.05.012.

DOI: 10.1016/j.csda.2011.05.012