响应变量删失时函数型部分线性分位数回归模型的估计

最近几年,函数型数据分析的理论和应用飞速发展.在许多实际应用里,响应变量往往存在随机右删失的情况.考虑利用函数型部分线性分位数回归模型来刻画函数型和标量预测量与右删失响应变量之间的关系.基于函数型主成分基函数来逼近未知的斜率函数,通过极小化逆概率加权分位数损失函数得到未知系数的估计量.文章的估计方法容易通过加权分位数回归程序实现.在一定的假设条件下,给出了有限维参数估计量的渐近正态性与斜率函数估计量的收敛速度.最后,通过模拟计算与应用实例证明了所提方法的有效性.     

参数单指标分位数自回归模型的诊断检验

本文研究参数单指标时间序列分位数自回归模型有效性的检验问题.当分位数回归变量的维数较大时,现有的检验方法将面临"维数灾难"问题.为了解决这个问题,本文基于残差经验过程,利用降维思想构造统计量,它有效地适应于参数单指标时间序列分位数自回归模型.本文提出Khmaladze鞅转换方法来替代经验过程,并构造检验统计量,证明所构造的检验统计量能够渐近收敛到分布自由的标准Brown运动.模拟研究和实际数据分析的结果表明,本文所提方法在参数单指标分位数自回归模型的检验中优于已有的检验方法.     

带相依辅助信息的分位数自回归模型的经验似然估计

分位数自回归模型作为一类常用的变系数时间序列模型,在理论研究和实际问题中都有广泛的应用.考虑到这类模型具有自回归的结构属性,数据采集过程中产生的额外信息,以相依辅助信息函数的形式被引入到模型系数的估计中来.该文应用经验似然方法得到了模型系数的估计量,得到了模型系数的估计量,并论证了其渐近正态性.基于渐近正态性的理论结果,进一步讨论了模型系数线性约束性问题的Wald检验统计量的渐近性质.数值模拟和实例数据分析的结果均表明,利用经验似然估计处理带相依辅助信息函数的方法较传统的分位数回归估计更有效.因而,一般常系数线性分位数回归模型在独立假设下的结果,被推广至具有相依结构的一类变系数模型中去.     

基于最小化复合分位损失函数的尺度参数估计和异质性检验

在进行回归分析时,对误差项离散程度的度量是一个重要话题.文章利用最小化复合分位损失的方法,对误差项的尺度参数进行估计,并证明估计量的大样本性质.进一步的研究表明:通过选取合适的分位数,能得到尺度参数的最优估计,并以此进行异质性检验.模拟结果表明,在重尾条件下所提出的方法有更高的精度.实际数据应用体现了该方法的良好性能.     

半参数平滑转换分位数自回归模型及应用

已有针对平滑转换自回归模型(STAR)的研究多是将转换函数设定为Logistic函数或指数函数形式,并在均值回归框架下获得模型的估计、检验及预测结果.文章基于重心权有理插值和分位数回归方法,构建一类新的半参数平滑转换分位数自回归模型,其主要特点表现在:第一,基于重心权有理插值方法构造的平滑转换函数,形式更加灵活自由,有效减少了模型误设的风险.第二,在分位数回归框架下,利用遗传算法获得新模型在不同分位点处的平滑转换自回归系数估计,比单纯的均值回归得到的信息更为丰富.数值模拟结果显示,新模型的平滑转换自回归系数估计在无偏性、有效性和一致性方面均具有较好表现.最后,将新模型应用于上证综指日收益率的动态趋势及预测研究,细致揭示了收益率序列在不同阶段、不同分位点处的非线性和异质性变化特征.     

带测量误差的线性分位数回归模型的工具变量估计（英文）

本文将工具变量法由研究带变量误差的均值回归模型推广到研究带变量误差的线性分位数回归模型.所得到的估计量是一致的且在一般条件下具有渐近正态分布.这种方法可行且易于操作.模拟研究表明该估计量在有限样本下性质表现非常好.最后这种方法被应用到实际问题,研究工资与教育程度之间的关系.     

纵向数据下半参数回归模型的统计分析

对于纵向数据下半参数回归模型,基于广义估计方程和一般权函数方法构造了模型中参数分量和非参数分量的估计.在适当的条件下证明了参数估计量具有渐近正态性,并得到了非参数回归函数估计量的最优收敛速度.通过模拟研究说明了所提出的估计量在有限样本下的精确性.     

固定效应空间自回归分位数模型的变量选择

本文研究了固定效应空间自回归分位数模型的变量选择问题.通过惩罚压缩相关参数,达到了同时识别空间效应、估计未知参数和选择解释变量的目的.此外,给出了变量选择的实现算法并证明了惩罚估计量的大样本性质.数值模拟和实例分析均表明了所提方法的优良表现.     

协变量缺失情形下的逆概率加权众数回归估计

数据缺失在实际应用中普遍存在,数据缺失会降低研究效率,导致参数估计有偏.在协变量随机缺失(MAR)的假定下,本文基于众数回归和逆概率加权估计方法对线性模型进行参数估计.该方法结合参数Logistic回归和非参数Nadaraya-Watson估计两种倾向得分估计方法,分别构建IPWM-L估计量和IPWM-NW估计量.模拟研究和实例分析表明,众数回归模型比均值回归模型更具稳健性,逆概率加权众数(IPWM)估计方法在缺失数据下表现出了更好的拟合效果,与IPWM-L估计量相比, IPWM-NW估计量更稳健.     

弱相依半参数回归模型的加权小波估计的渐近正态性

本文研究了半参数回归模型y_i=X'_iβ+g(t_i)+e_i,i=1,2,···,n,其中{e_i}为ψ-弱相依随机误差序列.利用小波估计的方法得到了参数、非参数的加权小波估计量.在相当一般的条件下,获得了这些小波估计量的渐近正态性,不仅推广了半参数回归模型的相应结果,而且在一定程度上统一了相依半参数回归模型的渐近正态性的理论.     

An effective method to reduce the computational complexity of composite quantile regression

In this article, we aim to reduce the computational complexity of the recently proposed composite quantile regression (CQR). We propose a new regression method called infinitely composite quantile regression (ICQR) to avoid the determination of the number of uniform quantile positions. Unlike the composite quantile regression, our proposed ICQR method allows combining continuous and infinite quantile positions. We show that the proposed ICQR criterion can be readily transformed into a linear programming problem. Furthermore, the computing time of the ICQR estimate is far less than that of the CQR, though it is slightly larger than that of the quantile regression. The oracle properties of the penalized ICQR are also provided. The simulations are conducted to compare different estimators. A real data analysis is used to illustrate the performance.     

Interquantile Shrinkage in Regression Models

Conventional analysis using quantile regression typically focuses on fitting the regression model at different quantiles separately. However, in situations where the quantile coefficients share some common feature, joint modeling of multiple quantiles to accommodate the commonality often leads to more efficient estimation. One example of common features is that a predictor may have a constant effect over one region of quantile levels but varying effects in other regions. To automatically perform estimation and detection of the interquantile commonality, we develop two penalization methods. When the quantile slope coefficients indeed do not change across quantile levels, the proposed methods will shrink the slopes toward constant and thus improve the estimation efficiency. We establish the oracle properties of the two proposed penalization methods. Through numerical investigations, we demonstrate that the proposed methods lead to estimations with competitive or higher efficiency than the standard quantile regression estimation in finite samples. Supplementary materials for the article are available online.     

Box-Cox变换用于分位数回归曲线相交问题的研究

将Box-Cox变换与分位数回归模型相结合(两阶段法),是分位数回归研究领域的一大进步。该法虽然两步都与分位数回归的检验函数紧密结合,但是由于没有利用分位数回归的优良性质,而是引入了中间参变量,因此增加了模型的累进误差,降低了模型精度。更重要的是,两阶段法没有对于分位数回归领域中普遍出现的分位数回归曲线的相交问题给出解决方法。针对这些问题,经研究应该首先确定Box-Cox变换的参数,避免模型中不确定因素的引入,然后对数据进行整体变换并结合分位数检验函数,直接利用分位数回归的优良性质,最终确定分位数回归模型的参数。实例证明,该方法提高了模型的精度,可以有效地解决分位数回归曲线的相交问题。     

Nonparametric quantile frontier estimation under shape restriction

This paper proposes a shape-restricted nonparametric quantile regression to estimate the τ-frontier, which acts as a benchmark for whether a decision making unit achieves top τ efficiency. This method adopts a two-step strategy: first, identifying fitted values that minimize an asymmetric absolute loss under the nondecreasing and concave shape restriction; second, constructing a nondecreasing and concave estimator that links these fitted values. This method makes no assumption on the error distribution and the functional form. Experimental results on some artificial data sets clearly demonstrate its superiority over the classical linear quantile regression. We also discuss how to enforce constraints to avoid quantile crossings between multiple estimated frontiers with different values of τ. Finally this paper shows that this method can be applied to estimate the production function when one has some prior knowledge about the error term.     

Quantile regression based on counting process approach under semi-competing risks data

In this paper, we investigate the quantile regression analysis for semi-competing risks data in which a non-terminal event may be dependently censored by a terminal event. The estimation of quantile regression parameters for the non-terminal event is complicated. We cannot make inference on the non-terminal event without extra assumptions. Thus, we handle this problem by assuming that the joint distribution of the terminal event and the non-terminal event follows a parametric copula model with unspecified marginal distributions. We use the stochastic property of the martingale method to estimate the quantile regression parameters under semi-competing risks data. We also prove the large sample properties of the proposed estimator, and introduce a model diagnostic approach to check model adequacy. From simulation results, it shows that the proposed estimator performs well. For illustration, we apply our proposed approach to analyze a real data.     

A generalized Newton algorithm for quantile regression models

This paper formulates the quadratic penalty function for the dual problem of the linear programming associated with the \(L_1\) constrained linear quantile regression model. We prove that the solution of the original linear programming can be obtained by minimizing the quadratic penalty function, with the formulas derived. The obtained quadratic penalty function has no constraint, thus could be minimized efficiently by a generalized Newton algorithm with Armijo step size. The resulting algorithm is easy to implement, without requiring any sophisticated optimization package other than a linear equation solver. The proposed approach can be generalized to the quantile regression model in reproducing kernel Hilbert space with slight modification. Extensive experiments on simulated data and real-world data show that, the proposed Newton quantile regression algorithms can achieve performance comparable to state-of-the-art.     

Unified Noncrossing Multiple Quantile Regressions Tree

In this article, we consider the estimation problem of a tree model for multiple conditional quantile functions of the response. Using the generalized, unbiased interaction detection and estimation algorithm, the quantile regression tree (QRT) method has been developed to construct a tree model for an individual quantile function. However, QRT produces different tree models across quantile levels because it estimates several QRT models separately. Furthermore, the estimated quantile functions from QRT often cross each other and consequently violate the basic properties of quantiles. This undesirable phenomenon reduces prediction accuracy and makes it difficult to interpret the resulting tree models. To overcome such limitations, we propose the unified noncrossing multiple quantile regressions tree (UNQRT) method, which constructs a common tree structure across all interesting quantile levels for better data visualization and model interpretation. Furthermore, the UNQRT estimates noncrossing multiple quantile functions simultaneously by enforcing noncrossing constraints, resulting in the improvement of prediction accuracy. The numerical results are presented to demonstrate the competitive performance of the proposed UNQRT over QRT. Supplementary materials for this article are available online.     

One-stage and two-stage DEA estimation of the effects of contextual variables

Two-stage data envelopment analysis (2-DEA) is commonly used in productive efficiency analysis to estimate the effects of operational conditions and practices on performance. In this method the DEA efficiency estimates are regressed on contextual variables representing the operational conditions. We re-examine the statistical properties of the 2-DEA estimator, and find that it is statistically consistent under more general conditions than earlier studies assume. We further show that the finite sample bias of DEA in the first stage carries over to the second stage regression, causing bias in the estimated coefficients of the contextual variables. This bias is particularly severe when the contextual variables are correlated with inputs. To address this shortcoming, we apply the result that DEA can be formulated as a constrained special case of the convex nonparametric least squares (CNLS) regression. Applying the CNLS formulation, we develop a new semi-nonparametric one-stage estimator for the coefficients of the contextual variables that directly incorporates contextual variables to the standard DEA problem. The proposed method is hence referred to as one-stage DEA (1-DEA). Evidence from Monte Carlo simulations suggests that the new 1-DEA estimator performs systematically better than the conventional 2-DEA estimator both in deterministic and noisy scenarios.     

多项式形的静态非线性投入产出模型解的存在性研究

经济系统时常表现明显的非线性特征，研究非线性投入产出具有重要的现实意义。在假定直接消耗系数是关于部门总产品的多项式函数的条件下，利用不动点理论求得在一定条件下，静态非线性投入产出模型的解的范围。由于多项式函数具有分段区间单词性的特点，使得多项式型的非线性投入产出模型突破了以往的研究中要求直接消耗系数仅仅是关于部门总产品的单调递减或单调递增的函数的限制，使建立非线性投入产出模型的条件更为宽松，更能描述现实的经济系统。     

Efficient estimation and variable selection for infinite variance autoregressive models

In this paper, a self-weighted composite quantile regression estimation procedure is developed to estimate unknown parameter in an infinite variance autoregressive (IVAR) model. The proposed estimator is asymptotically normal and more efficient than a single quantile regression estimator. At the same time, the adaptive least absolute shrinkage and selection operator (LASSO) for variable selection are also suggested. We show that the adaptive LASSO based on the self-weighted composite quantile regression enjoys the oracle properties. Simulation studies and a real data example are conducted to examine the performance of the proposed approaches.