摘要
假定随机误差分布来自具有重尾特征的scale mixtures of normal分布族,运用贝叶斯方法研究了函数型线性回归模型的稳健性估计,其中模型的响应变量为标量,解释变量为函数型变量.数值模拟结果表明:当响应变量的观测数据存在离群值时,建立的方法得到的模型参数的估计,要优于正态分布假定下的模型参数的估计.
It gives Bayesian robust estimations of functional linear regression models, where the response is a scalar and the predictor is functional, by applying a class of heavy-tailed scale mixtures of normal distributions for random measurement errors. The numerical results showed that: the estimations of the model parameters using scale mixtures of normal distributions outperform the estimations using normal distributions.
引文
[1] Ramsay J O, and Silverman B W. Functional Data Analysis[M](2nd ed.). Springer, New York,USA, 2005.
[2] Ramsay J O, and Silverman B W.函数型数据分析[M](第二版影印版).科学出版社,北京,2006.
[3] Ferraty F, and Vieu P. Nonparametric Functional Data Analysis:Theory and Practice[M].Springer New York USA, 2006.
[4] Koenker R, and Bassett G. JR. Regression quantiles[J]. Econometrica, 1978, 46:33-50.
[5] Cardot H, Crambes C, and Sarda P. Quantile regression when the covariates are functions[J].Journal of Nonparametric Statistics, 2005, 17:841-856.
[6] Ferraty F, Rabhi A, and Vieu P. Conditional quantiles for dependent functional data with application to the climatic El Nino phenomenon[J]. Sankhya, 2005, 67:378-398.
[7] Kato K. Estimation in functional linear quantile regression[J]. The Annals of Statistics, 2012,40(6):3108-3136.
[8] Lu Y, Du J, and Sun Z. Functional partially linear quantile regression model[J]. Metrika, 2014,77:317-332.
[9] Tang Q, and Kong L. Quantile regression in functional linear semiparametric model[J]. Statistics,2017, 51(6):1342-1358.
[10] Zhou J, Du J, and Sun Z. M-estimation for partially functional linear regression model based on splines[J]. Communications in Statistics, 2015, 45(21):6436-6446.
[11] Boente G, and Vahnovan A. Robust estimators in semi-functional partial linear regression models[J]. Journal of Multivariate Analysis, 2017, 154:59-84.
[12] Andrews D F, and Mallows C L. Scale mixtures of normal distributions[J]. Journal of the Royal Statistical Society, Series B, 1974, 36:99-102.
[13] Liu C. Bayesian robust multivariate linear regression with incomplete data[J]. Journal of the American Statistical Association, 1996, 91:1219-1227.
[14] Lange K, and Sinsheimer J. Normal/independent distributions and their applications in robust regression[J]. Journal of Computational and Graphical Statistics, 1993, 2:175-198.
[15] Choy S T B, and Smith A F M. Hierarchical models with scale mixtures of normal distributions[J]. Test, 1997, 6:205-221.
[16] Rosa G J M, Gianola D, and Padovani C R. Bayesian longitudinal data analysis with mixed models and thick-tailed distributions using MCMC[J]. Journal of Applied Statistics 2004, 31:855-873.
[17] Meza C, Osorio F, and la Cruz R D. Estimation in nonlinear mixed-effects models using heavytailed distributions[J]. Statistics and Computing 2012, 22:121-139.
[18] De la Cruza R. Bayesian analysis for nonlinear mixed-effects models under heavy-tailed distributions[J]. Pharmaceutical Statistics 2014, 13:81-93.