Adaptive stochastic gradient boosting tree with composite criterion

ABSTRACT

In this paper, we propose an adaptive stochastic gradient boosting tree for classification studies with imbalanced data. The adjustment of cost-sensitivity and the predictive threshold are integrated together with a composite criterion into the original stochastic gradient boosting tree to deal with the issues of the imbalanced data structure. Numerical study shows that the proposed method can significantly enhance the classification accuracy for the minority class with only a small loss in the true negative rate for the majority class. We discuss the relation of the cost-sensitivity to the threshold manipulation using simulations. An illustrative example of the analysis of suboptimal health-state data in traditional Chinese medicine is discussed.     

Dimension reduction via local rank regression

Outer product of gradients (OPG) achieves dimension reduction via estimating the gradients of the regression function. In this paper, we propose two novel OPG estimators via local rank regression: the rank OPG estimator and the Walsh-average OPG estimator. Both proposals guard against a wide range of error distributions, and are safe alternatives to existing OPG estimators based on local linear regression or local L1 regression. The effectiveness of the new proposals are demonstrated via extensive numerical studies.     

近代广东商业人群在广西的分布面及其梯度差考详——粤港商业经济对广西经济辐射研究之二

文章从中国近代史上城镇经济主体——商业经济的角度切入,运用各类历史资料,分区逐县地确证广东商业人群以及所造成的“无东不成市”格局在全广西的分布范围,并揭示其在广西不同地域的程度等差和梯度分布状态。从而,以原始的生活场景原生态地呈现出粤港澳地区对广西经济辐射由东而西逐步推进、由城市而圩市逐层次渗透的基本节奏和客观规律。     

城乡二元结构下农村公共产品的供给困境与化解

在城乡二元结构下,与城市相比,农村公共产品供给呈现出总量不足、结构失调、供需矛盾、主体失衡的困境。造成农村公共产品供给困境的主要制度根源在于城乡二元结构,城乡二元结构既是造成农村公共产品供给事权与财权不对称的重要制度环境,又是导致市场和社会组织等多元化供给主体缺失的重要因素。因此,破除城乡二元结构,完善农村公共产品供给体制,就成为化解农村公共产品供给困境的关键所在。     

位移边界对2D-FGM板热应力影响研究

考虑不同的位移边界条件,建立了常物性2D-FGM板研究模型,并采用有限元法,得到其冷却瞬态热应力场分布图形。位移边界条件对常物性2D-FGM平面区域冷却瞬态热应力分布影响颇大,在简支和一端固定情况下,2D-FGM板自由边界上的热应力σ_x=0;两端固定情况下,其左右两边界均形成上中下三个应力聚集,应力分布形状和数值都发生了巨大的变化;将四周均设置呈固定约束后,热应力数值增长幅度较两端固定情况时更为明显。此结果为该材料的设计、制备提供了准确的理论计算依据。     

How Close Is Close Enough: Public Perceptions of the Risks of Industrial Facilities

Public opinion poll data have consistently shown that the proportion of respondents who are willing to have a nuclear power plant in their own community is smaller than the proportion who agree that more nuclear plants should be built in this country. Respondents' judgments of the minimum safe distance from each of eight hazardous facilities confirmed that this finding results from perceived risk gradients that differ by facility (e.g., nuclear vs. natural gas power plants) and social group (e.g., chemical engineers vs. environmentalists) but are relatively stable over time. Ratings of the facilities on thirteen perceived risk dimensions were used to determine whether any of the dimensions could explain the distance data. Because the rank order of the facilities with respect to acceptable distance was very similar to the rank order on a number of the perceived risk dimensions, it is difficult to determine which of the latter is the critical determinant of acceptable distance if, indeed, there is only one. There were, however, a number of reversals of rank order that indicate that the respondents had a differentiated view of technological risk. Finally, data from this and other studies were interpreted as suggesting that perceived lack of any other form of personal control over risk exposure may be an important factor in stimulating public opposition to the siting of hazardous facilities.     

新吉坪隧道斜井进正洞挑顶施工技术研究

沪昆铁路客运专线新吉坪隧道所处地质条件复杂，区域地质作用剧烈，碳酸岩分布广泛，不良地质特别发育。针对斜井进正洞施工，提出了纵向爬坡导坑法。并以2#斜井为例，介绍了纵向爬坡导坑法施工工艺和斜井进正洞挑顶施工的控制要点，以及在施工过程中的控制量测方法，为新吉坪隧道斜井进正洞挑顶施工提供了技术指导，为类似工程施工提供参考。     

Urban tree cover: an ecological perspective

Analysis of urban tree cover is generally limited to inventories of tree structure and composition on public lands. This approach provided valuable information for resource management. However, it does not account for all tree cover within an urban landscape, thus providing insufficient information on ecological patterns and processes. We propose evaluating tree cover for an entire urban area that is based on patch dynamics. Treed patches are classified by their origin, structure, and management intensity. A patch approach enables ecologists to evaluate ecological patterns and processes for the entire urban landscape and to examine how social patterns influence these ecological patterns and processes.     

Nonparametric multiple expectile regression via ER-Boost

Expectile regression [Newey W, Powell J. Asymmetric least squares estimation and testing, Econometrica. 1987;55:819–847] is a nice tool for estimating the conditional expectiles of a response variable given a set of covariates. Expectile regression at 50% level is the classical conditional mean regression. In many real applications having multiple expectiles at different levels provides a more complete picture of the conditional distribution of the response variable. Multiple linear expectile regression model has been well studied [Newey W, Powell J. Asymmetric least squares estimation and testing, Econometrica. 1987;55:819–847; Efron B. Regression percentiles using asymmetric squared error loss, Stat Sin. 1991;1(93):125.], but it can be too restrictive for many real applications. In this paper, we derive a regression tree-based gradient boosting estimator for nonparametric multiple expectile regression. The new estimator, referred to as ER-Boost, is implemented in an R package erboost publicly available at http://cran.r-project.org/web/packages/erboost/index.html. We use two homoscedastic/heteroscedastic random-function-generator models in simulation to show the high predictive accuracy of ER-Boost. As an application, we apply ER-Boost to analyse North Carolina County crime data. From the nonparametric expectile regression analysis of this dataset, we draw several interesting conclusions that are consistent with the previous study using the economic model of crime. This real data example also provides a good demonstration of some nice features of ER-Boost, such as its ability to handle different types of covariates and its model interpretation tools.     

区域经济发展模式的特点与选择

我国经济理论界和实际工作,根据国情特点,联系各地不平衡发展的现状,先后提出了梯度推进发展,点轴渐进发展,区域经济协调发展,城市圈域经济发展和网状交织发展等区域经济发展模式。它们有不同的特色和优势,但也存在某些不足。从长远看,比较起来,网状交织发展模式更有利于提高区域经济的运行质量,宜于进入工业化中后期阶段的地区采用。