统计与数据科学学院
------------------------------------------------------------------------------------------
学术讲座
地点：统计与数据科学学院会议室
时间：2020年10月21日 14:00
邀请人：韩东啸 博士
报告人及题目：
（1）陈坤
个人简介：陈坤，西南财经大学统计学院副教授，香港中文大学博士。陈坤博士在《Insurance: Mathematics and Economics》，《Journal of Time Series Analysis》，《Electronic Journal of Statistics》，《Canadian Journal of Statistics》，《Journal of Statistical Planning and Inference》等国际知名学术期刊上发表过多篇文章。主要研究领域包括时间序列分析，空间统计，函数型数据分析，金融统计等。
报告题目：Subgroup analysis of zero-Inflated Poisson regression model with applications to insurance data.
摘要：Customized personal rate offering is of growing importance in the insurance industry. To achieve this, an important step is to identify subgroups of insureds from the corresponding heterogeneous claim frequency data. In this paper, a penalized Poisson regression approach for subgroup analysis in claim frequency data is proposed. Subjects are assumed to follow a zero-inflated Poisson regression model with group-specific intercepts, which capture group characteristics of claim frequency. A penalized likelihood function is derived and optimized to identify the group-specific intercepts and effects of individual covariates. To handle the challenges arising from the optimization of the penalized likelihood function, an alternating direction method of multipliers algorithm is developed and its convergence is established. Simulations studies and real applications are provided for illustrations.
（2）郝美玲
个人简介：郝美玲，对外经济贸易大学副教授，香港理工大学博士，多伦多大学附属医院玛格丽特公主癌症研究中心博士后。郝美玲博士在《Journal of the American Statistical Association》，《Statistica Sinica》，《Genetic Epidemiology》等国际知名学术期刊上发表过多篇文章。主要研究领域包括高维数据分析，生物统计，非参数统计，生物信息工程等。
报告题目：Functional additive hazards model with application to COVID-19 data.
摘要: We propose a new functional additive hazards model to investigate potential effects of functional and scalar predictors on mortality risks, and develop a penalized least squares estimation approach for model parameters. The consistency, the convergence rate and the asymptotic distribution of the resulting estimator are established. In particular, deriving the asymptotic joint distribution of the infinite-dimensional and the finite-dimensional estimators poses great challenges. To tackle this problem, we design a framework of the Sobolev space equipped with a proper inner product and obtain a joint Bahadur representation of the estimators of functional and scalar parameters in the space. Using this key result, we further establish the asymptotic joint normality of the proposed estimators. Our simulation studies demonstrate that the proposed estimation procedure performs well. For illustration, we apply the proposed method to the COVID-19 data that motivated this research. The analysis results provide evidence to support the claim that minimizing community interactions indeed reduces mortality risks induced by COVID-19.
（3）谢锦瀚
个人简介：谢锦瀚，云南大学博士，香港中文大学博士后。谢锦瀚博士在《Journal of the American Statistical Association》，《Statistica Sinica》，《Computational Statistics and Data Analysis》等国际知名学术期刊上发表过多篇文章。主要研究领域包括高维数据分析，模型平均，缺失数据等。
报告题目：A model-averaging method for high-dimensional regression with missing responses at random.
摘要: This study considers the ultrahigh-dimensional prediction problem in the presence of responses missing at random. A two-step model-averaging procedure is proposed to improve the prediction accuracy of the conditional mean of the response variable. The first step specifies several candidate models, each with low-dimensional predictors. To implement this step, a new feature-screening method is developed to distinguish between the active and inactive predictors. The method uses the multiple-imputation sure independence screening (MI-SIS) procedure, and candidate models are formed by grouping covariates with similar size MI-SIS values. The second step develops a new criterion to find the optimal weights for averaging a set of candidate models using weighted delete-one cross-validation (WDCV). Under some regularity conditions, we show that the proposed screening statistic enjoys the ranking consistency property, and that the WDCV criterion asymptotically achieves the lowest possible prediction loss. Simulation studies and an example demonstrate the proposed methodology.
（4）张政
个人简介：张政，中国人民大学统计与大数据研究院助理教授，香港中文大学博士。张政博士在《Journal of the Royal Statistical Society: Series B》，《Stochastic Processes and their Applications》，《Statistica Sinica》，《Journal of Multivariate Analysis》，《Journal of Statistical Planning and Inference》，等国际知名学术期刊上发表过多篇文章。主要研究领域包括因果推断，缺失数据，半参/非参数统计，随机分析等。
报告题目：A simple and efficient estimation of average treatment effects in models with unmeasured confounders.
摘要：This paper presents a simple and efficient estimation of the average treatment effect(ATE) and local average treatment effect (LATE) in models with unmeasured confounders. In contrast with the existing studies which estimate some unknown functionals in the influence function either parametrically or semi parametrically, we do not model the influence function. Instead we apply the calibration method to a growing number of moment restrictions to estimate the weighting functions non parametrically and then estimate ATE and LATE by plugging in. The calibration method is similar to the covariate-balancing method in that both methods exploit the moment restrictions. The difference is that the calibration method imposes the sample analogue of the moment restrictions, which is the key for efficient estimation of ATE and LATE. A simulation study reveals that our estimators have good finite sample performance and outperform the existing alternatives. An application to the empirical analysis of return to education illustrates the practical value of the proposed method.