Frechet single index models for object response regression

被引:8
|
作者
Ghosal, Aritra [1 ]
Meiring, Wendy [1 ]
Petersen, Alexander [2 ]
机构
[1] Univ Calif Santa Barbara, Dept Stat & Appl Probabil, Santa Barbara, CA 93106 USA
[2] Brigham Young Univ, Dept Stat, Provo, UT USA
来源
ELECTRONIC JOURNAL OF STATISTICS | 2023年 / 17卷 / 01期
关键词
Frechet regression; single-index model; random objects; local smoothing; STRONG UNIFORM CONSISTENCY; EXTRINSIC SAMPLE MEANS; LEAST-SQUARES; DENSITY; MORTALITY; MANIFOLDS; SPACE; WEAK;
D O I
10.1214/23-EJS2120
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
With the increasing availability of non-Euclidean data objects, statisticians are faced with the task of developing appropriate statistical methods for their analysis. For regression models in which the predictors lie in R-p and the response variables are situated in a metric space, conditional Frechet means can be used to define the Frechet regression function. Global and local Frechet methods have recently been developed for modeling and estimating this regression function as extensions of multiple and local linear regression, respectively. This paper expands on these methodologies by proposing the Frechet single index model, in which the Frechet regression function is assumed to depend only on a scalar projection of the multivariate predictor. Estimation is performed by combining local Frechet along with M-estimation to estimate both the coefficient vector and the underlying regression function, and these estimators are shown to be consistent. The method is illustrated by simulations for response objects on the surface of the unit sphere and through an analysis of human mortality data in which lifetable data are represented by distributions of age-of-death, viewed as elements of the Wasserstein space of distributions.
引用
收藏
页码:1074 / 1112
页数:39
相关论文
共 50 条
  • [41] Effects of measurement error on a class of single-index varying coefficient regression models
    Shi, Jianhong
    Yang, Qian
    Li, Xiongya
    Song, Weixing
    COMPUTATIONAL STATISTICS, 2017, 32 (03) : 977 - 1001
  • [42] Empirical Likelihood for Single-Index Regression Models under Negatively Associated Errors
    Lin, Zheng-Yan
    Wang, Ran
    COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 2015, 44 (09) : 1854 - 1868
  • [43] Semi-parametric single-index two-part regression models
    Zhou, XH
    Liang, H
    COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2006, 50 (05) : 1378 - 1390
  • [44] Estimation and testing for varying-coefficient single-index quantile regression models
    Ding, Hui
    Yao, Mei
    Zhang, Riquan
    Zhang, Zhenglong
    Zhu, Hanbing
    JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2025, 239
  • [45] Extreme quantile regression for tail single-index varying-coefficient models
    Wang, Yingjie
    Liu, Xinsheng
    COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 2023, 52 (09) : 2860 - 2881
  • [46] Quantile regression and variable selection for single-index varying-coefficient models
    Yang, Jing
    Yang, Hu
    COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2017, 46 (06) : 4637 - 4653
  • [47] Semi-parametric single-index predictive regression models with cointegrated regressors
    Zhou, Weilun
    Gao, Jiti
    Harris, David
    Kew, Hsein
    JOURNAL OF ECONOMETRICS, 2024, 238 (01)
  • [48] Effects of measurement error on a class of single-index varying coefficient regression models
    Jianhong Shi
    Qian Yang
    Xiongya Li
    Weixing Song
    Computational Statistics, 2017, 32 : 977 - 1001
  • [49] Nonconcave penalized inverse regression in single-index models with high dimensional predictors
    Zhu, Li-Ping
    Zhu, Li-Xing
    JOURNAL OF MULTIVARIATE ANALYSIS, 2009, 100 (05) : 862 - 875
  • [50] Single-index quantile regression
    Wu, Tracy Z.
    Yu, Keming
    Yu, Yan
    JOURNAL OF MULTIVARIATE ANALYSIS, 2010, 101 (07) : 1607 - 1621