Nonparametric Regression with Measurement Error in Banach Spaces
本文提出一种适用于Banach空间值预测变量的非参数回归框架，解决在无内积结构及测量误差情况下的估计问题。
0 citationsRead paper
Selected work
Representative Papers
Nonparametric Estimation of Event-Free Survival for Data with Left-Truncated Death and Intermittently Assessed Nonfatal Events
该研究提出了一种核平滑方法，用于在存在左截断死亡和间歇评估非致命事件的数据中非参数估计无事件生存率。
0 citationsRead paper
Covariate-Adjusted Functional Principal Components Analysis for Modeling Hazard Rates of Physical Activity in the US Population
Traditional summary metrics struggle to capture the heterogeneity in individual physical activity intensity distributions. This study proposes a novel approach that integrates wrist-worn accelerometer-derived individual activity intensity hazard functions with functional data analysis. By employing nonparametric modeling, log-transformation, and covariate-adjusted functional principal component analysis (FPCA), the method uncovers dominant modes of variation in population-level physical activity patterns. Moving beyond mean-based summaries, this framework offers a more flexible and interpretable characterization of heterogeneity. Applied to large-scale NHANES data, the approach successfully identifies significant differences in high- and low-intensity activity profiles across population subgroups, demonstrating its superior capability in describing and comparing physical activity distributions.
0 citationsRead paper
Recent publications
Latest Papers
Nonparametric Regression with Measurement Error in Banach Spaces
本文提出一种适用于Banach空间值预测变量的非参数回归框架，解决在无内积结构及测量误差情况下的估计问题。
0 citationsRead paper
Nonparametric Estimation of Event-Free Survival for Data with Left-Truncated Death and Intermittently Assessed Nonfatal Events
该研究提出了一种核平滑方法，用于在存在左截断死亡和间歇评估非致命事件的数据中非参数估计无事件生存率。
0 citationsRead paper
Covariate-Adjusted Functional Principal Components Analysis for Modeling Hazard Rates of Physical Activity in the US Population
Traditional summary metrics struggle to capture the heterogeneity in individual physical activity intensity distributions. This study proposes a novel approach that integrates wrist-worn accelerometer-derived individual activity intensity hazard functions with functional data analysis. By employing nonparametric modeling, log-transformation, and covariate-adjusted functional principal component analysis (FPCA), the method uncovers dominant modes of variation in population-level physical activity patterns. Moving beyond mean-based summaries, this framework offers a more flexible and interpretable characterization of heterogeneity. Applied to large-scale NHANES data, the approach successfully identifies significant differences in high- and low-intensity activity profiles across population subgroups, demonstrating its superior capability in describing and comparing physical activity distributions.
0 citationsRead paper