Shrinkage Estimation of the Varying Coefficient Model

By Hansheng Wang, Yingcun Xia. 2009

306 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/e87f9b613d1012126a2b3fc62bc861e8054dd628

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Heng Lian
2. Hua Liang
3. Byeong U. Park
4. Heng Lian
5. Jian Huang
6. Fengrong Wei
7. Eun Ryung Lee
8. Jianyi Yang

In Statistics and Probability

1. Heng Lian
2. Byeong U. Park
3. Hua Liang
4. Enno Mammen
5. Eun Ryung Lee
6. Jianqing Fan
7. Young K. Lee
8. Tao Hu

Citations in the corpus

• A varying-coefficient panel data model with fixed effects: Theory and an application to US commercial banks (2017, unknown): “As in Wang and Xia (2009), we use the following data-driven regularizer weight: τ = τ̃ ( ‖b̃1‖−1, . . . , ‖b̃q‖−1 )′ , (2.15) where τ̃ is a scalar, b̃s is the s th column of the unregularized estimator B̃, and B̃ is obtained from (2.14) by simply choosing τ1 = · · · = τq = 0.”
• Identification of Partially Linear Structure in Additive Models with an Application to Gene Expression Prediction from Sequences (2012, methodology): “Furthermore, both SCAD penalty and adaptive lasso (Fan and Li, 2001; Zou, 2006; Zhang and Lu, 2007; Wang, Li, and Huang, 2008; Wang and Xia, 2009; Xie and Huang, 2009; Huang, Horowitz, and Wei, 2010) were proposed to address the consistency issues of the lasso.”
• Circuit Theory and Model-Based Inference for Landscape Connectivity (2013, methodology): “For comparison, we also employed a correlation analysis, the most common existing method in landscape genetics studies (e.g., Cushman et al. 2006; Cushman, McKelvey, and Schwartz 2009; Wang and Xia 2009; Shirk et al. 2010).”
• A semiparametric spatial dynamic model (2014, background): “[19, 20]; Wang and Xia [17]; and Tao and Xia [16]).”
• VARIABLE SELECTION IN NONPARAMETRIC ADDITIVE MODELS (2010, methodology): “For example, Wang, Chen and Li (2007) and Wang and Xia (2008) considered the use of group Lasso and SCAD methods for model selection and estimation in varying coefficient models with a fixed number of coefficients and covariates.”
• Varying Coefficient Regression Models: A Review and New Developments (2015, methodology): “Wang & Xia (2009) claimed that the method correctly identifies I0 with probability tending to one, and that the resulting estimators are as efficient as the oracle estimators, which uses the knowledge of I0.”
• ADAPTIVE SEMI-VARYING COEFFICIENT MODEL SELECTION (2012, methodology): “To evaluate estimation accuracy, we considered the relative estimation error (REE, Wang and Xia (2009)) REE = 100× ∑n i=1 ∑p j=1 |β̂λ,j(Zi)− β0j(Zi)|∑n i=1 ∑p j=1 |β̄j(Zi)− β0j(Zi)| , where β̄j(·) is either the unpenalized estimator or the oracle estimator.”
• Variable selection for high-dimensional generalized varying-coefficient models (2012, methodology): “For the ordinary varying-coefficients models with quadratic loss function, Wang, Li, and Huang (2008) and Wang and Xia (2009) each proposed a group penalization method in the fixed p case, and Wei, Huang, and Li (2011) recently extended this work to the case of diverging p.”
• Model Selection via Bayesian Information Criterion for Quantile Regression Models (2014, methodology): “Additionally, it seems to be possible to consider the extension from the BIC for parametric quantile regression using the effective sample size nh in kernel smoothing, instead of the original sample size n, as Wang and Xia (2009) did in varying coefficient mean regression models. In our simulations,”
• A UNIFIED VARIABLE SELECTION APPROACH FOR VARYING COEFFICIENT MODELS (2012, methodology): “For least squares regression, edf1 is defined as the total number of varying and nonzero constant coefficients (Wang and Xia (2009)).”
• GENERALIZED ADDITIVE PARTIAL LINEAR MODELS WITH HIGH-DIMENSIONAL COVARIATES (2013, background): “In the literature, both generalized cross-validation (GCV) and Bayesian information criterion (BIC) have been advocated for selection of tuning parameters for penalized sparse models (Huang et al., 2010; Wang, Li, and Tsai, 2007; Wang and Xia, 2009; Wang et al., 2011).”
• Shrinkage estimation and selection for multiple functional regression (2011, methodology): “, 2008) where the authors used basis expansion approach for estimation combined with group SCAD penalty on coefficients, and the work (Wang and Xia, 2009) where group Lasso penalty is applied directly to smooth functions evaluated at sampled points.”