Oracle Inequalities for Convex Loss Functions with Nonlinear Targets

This article considers penalized empirical loss minimization of convex loss functions with unknown target functions. Using the elastic net penalty, of which the Least Absolute Shrinkage and Selection Operator (Lasso) is a special case, we establish a finite sample oracle inequality which bounds the loss of our estimator from above with high probability. If the unknown target is linear, this inequality also provides an upper bound of the estimation error of the estimated parameter vector. Next, we use the non-asymptotic results to show that the excess loss of our estimator is asymptotically of the same order as that of the oracle. If the target is linear, we give sufficient conditions for consistency of the estimated parameter vector. We briefly discuss how a thresholded version of our estimator can be used to perform consistent variable selection. We give two examples of loss functions covered by our framework.     

Conditional VAR and Expected Shortfall: A New Functional Approach

We estimate two well-known risk measures, the value-at-risk (VAR) and the expected shortfall, conditionally to a functional variable (i.e., a random variable valued in some semi(pseudo)-metric space). We use nonparametric kernel estimation for constructing estimators of these quantities, under general dependence conditions. Theoretical properties are stated whereas practical aspects are illustrated on simulated data: nonlinear functional and GARCH(1,1) models. Some ideas on bandwidth selection using bootstrap are introduced. Finally, an empirical example is given through data of the S&P 500 time series.     

Functional Partial Linear Single‐index Model

This paper deals with the problem of predicting the real‐valued response variable using explanatory variables containing both multivariate random variable and random curve. The proposed functional partial linear single‐index model treats the multivariate random variable as linear part and the random curve as functional single‐index part, respectively. To estimate the non‐parametric link function, the functional single‐index and the parameters in the linear part, a two‐stage estimation procedure is proposed. Compared with existing semi‐parametric methods, the proposed approach requires no initial estimation and iteration. Asymptotical properties are established for both the parameters in the linear part and the functional single‐index. The convergence rate for the non‐parametric link function is also given. In addition, asymptotical normality of the error variance is obtained that facilitates the construction of confidence region and hypothesis testing for the unknown parameter. Numerical experiments including simulation studies and a real‐data analysis are conducted to evaluate the empirical performance of the proposed method.     

中国儿童基本机会多维不平等的综合测度研究

遵循国际规范,基于世界银行用于测度单一机会不平等的人类机会指数(Human Opportunity Index,HOI),借鉴最新的综合测度方法和多维机会指数(Multidimensional Human Opportunity Index,MHOI),首次对中国儿童的教育、医疗保险、安全饮水、卫生设施及家庭用电5种基本机会不平等进行综合测度。实证结果显示,2011年,中国7¹2岁儿童最少享有2种基本机会,所有基本机会全部享有的儿童不到60%。在所考虑的基本机会中,儿童卫生设施的覆盖率、公平性和HOI是最低的,其次是安全饮水;卫生设施和安全饮水的HOI表现出较大的地区和城乡差异。从而导致MHOI较大的地区和城乡差异;动态分析结果表明,近10多年来,中国儿童各种基本机会的覆盖率和公平性都得到了持续的改善,但各基本机会的HOI和MHOI仍存在较大的改善空间。这些量化结论具有一定的政策指导意义。     

煤炭大数据指数编制及经验模态分解模型研究

基于开放性数据源、连续观测昨多变量数据编制的大数据指数,与传统的统计调查指数存在的差异不仅在于数据本身的无限扩张,而且在于编制方法以及分解研究的规则、模型方面的差异。在大数据背景下,率先尝试性地提出大数据指数的定义和数据假设,将"互联网大数据指数"引入煤炭交易价格指数综合编制太原煤炭交易大数据指数,从而反映煤炭价格的变动趋势;导入经验模态分解模型,对所编制的煤炭大数据指数进行分解研究,尝试比较与传统的统计调查指数的差异。研究表明:新编制的煤炭价格大数据指数要比太原煤炭交易价格指数更为敏感和迅速,能更好地反映煤炭价格的变动趋势。随着"互联网+"和大数据战略的逐渐普及,基于互联网大数据编制的综合指数会影响到更多领域,将成为经济管理和社会发展各个领域的晴雨表和指示器;与传统统计调查指数逐步融合、互补或者升级,成为宏观经济大数据指数的重要组成部分。     

Cultural capital or relative risk aversion? Two mechanisms for educational inequality compared1

In this paper we empirically examined two explanatory mechanisms for educational inequality: cultural reproduction and relative risk aversion, using survey data taken from secondary school pupils in Amsterdam. Cultural reproduction theory seeks to explain class variations in schooling by cultural differences between social classes. Relative risk aversion theory argues that educational inequalities can be understood by between-class variation in the necessity of pursuing education at branching points in order to avoid downward mobility. We showed that class variations in early demonstrated ability are for a substantial part cultural: cultural capital - measured by parental involvement in highbrow culture - affected school performance at the primary and secondary level. However, relative risk aversion - operationalized by being concerned with downward mobility - strongly affects schooling ambitions, whereas cultural capital had no effect. Thus, we conclude that 'primary effects' of social origin on schooling outcomes are manifested through cultural capital and not through relative risk aversion (in addition to other potential sources of class variations such as genetics). Relative risk aversion, and not cultural capital, affects schooling ambitions, which is relevant for our understanding of secondary effects.     

Driven by Inequalities: Exploring the Resurgence of Domestic Work in U.S. Cities

Domestic work, once the most common occupation for women around the globe, was thought to be well on its way to extinction at the end of the twentieth century. However, in the 1980s and 1990s, evidence began to appear that domestic work was in many places again becoming a growth occupation. My goal in this article is to examine the factors related to the recent expansion of domestic work in countries in the Global North, using the United States as a case study. I draw on U.S. Census data to document the resurgence of domestic work both nationally and in many large cities across the country, and then use multivariate analysis to compare rates of domestic work across these cities. The results indicate that rates of domestic work are highly related to variables measuring structural inequalities (racialization of the labor force, immigration, and economic polarization), while showing little relationship with variables measuring unmet care needs (care dependency ratios, female/maternal labor force participation, and availability of institutional care options). These findings underline the urgency of providing protections to domestic workers and point to the need for scholarship that better theorizes the relationships among unpaid care and different forms of paid care.     

Will You Die for Your Country? Workplace Death in an Era of Mass Incarceration

People with criminal records in the United States continue to face limited employment opportunities due to social stigma and legal barriers. In contrast to the civilian sector, the military conducts a “whole person” evaluation to screen potential recruits and regularly hires people with felony and misdemeanor records. Moreover, evidence suggests that the military serves as a socially integrative institution and may facilitate desistance from future crimes. However, critics argue that the military exacerbates inequalities by subjecting marginalized communities to the unequal burden of service. Using the data obtained from the Army, we examine the relative risks of combat exposure and casualties between enlisted soldiers with and without criminal records who joined between 2002 and 2009. The results suggest that soldiers with felony and misdemeanor records are more likely to be assigned to combat occupations than those without criminal records. We also find that among soldiers assigned to positions with low combat exposure, ex-offenders face a higher risk of death compared to those without criminal records. Findings do not dispute the idea that the military facilitates desistance from future crimes and provides second chances to people with criminal records, but reaffirm the fact that military service costs lives and limbs.