A mathematical programming approach for improving the robustness of least sum of absolute deviations regression

This paper discusses a novel application of mathematical programming techniques to a regression problem. While least squares regression techniques have been used for a long time, it is known that their robustness properties are not desirable. Specifically, the estimators are known to be too sensitive to data contamination. In this paper we examine regressions based on Least‐sum of Absolute Deviations (LAD) and show that the robustness of the estimator can be improved significantly through a judicious choice of weights. The problem of finding optimum weights is formulated as a nonlinear mixed integer program, which is too difficult to solve exactly in general. We demonstrate that our problem is equivalent to a mathematical program with a single functional constraint resembling the knapsack problem and then solve it for a special case. We then generalize this solution to general regression designs. Furthermore, we provide an efficient algorithm to solve the general nonlinear, mixed integer programming problem when the number of predictors is small. We show the efficacy of the weighted LAD estimator using numerical examples. © 2006 Wiley Periodicals, Inc. Naval Research Logistics, 2006     

Emulating US Counterinsurgency Doctrine: Barriers for Developing Country Forces,Evidence from Peru

ABSTRACT

Recent US advances in counterinsurgency doctrine have been adopted by developing country armed forces. Nevertheless, no systematic study has examined the barriers they face to implementing highly involved counterinsurgency strategy. Tracing the evolution of Peruvian doctrine demonstrates that Peru was able to quickly improve the unity of effort, intelligence capacity, and military basing to meet the demands of a population-centric hearts-and-minds approach to counterinsurgency. Nevertheless, the limited tactical initiative and flexibility of Peruvian forces remains a challenge. The Peruvian experience is instructive for other militaries undergoing similar transitions. However, given the diversity of insurgent conflicts, this doctrine is not universally appropriate.     

The impact of Syrian refugees on the labor market in neighboring countries: empirical evidence from Jordan

This paper analyzes time-sensitive data on a humanitarian crisis in the Middle East. It aims to assess the impact of the steep influx of Syrian refugees into Jordan on the country’s labor market since the onset of the conflict in Syria (March 2011). As of August 2014, nearly three million registered Syrians have sought refuge in neighboring countries (Lebanon, Jordan, Iraq, and Turkey), according to the United Nations High Commissioner for Refugees. Jordan and Lebanon are hosting the majority of them. This paper utilizes data regarding unemployment rates, employment rates, labor force participation, the number of refugees, and economic activity at the level of governorates. The vector autoregressive methodology is used to examine time series data from the most affected governorates in Jordan. The empirical results of Granger causality tests and impulse response functions show that there is no relationship between the influx of Syrian refugees and the Jordanian labor market. Our results are verified through a set of robustness checks.     

Understanding armed groups and party politics

Abstract

Political parties with an armed history are not unusual, yet how these groups function in politics after the transition has largely been ignored. This special issue examines armed groups in party politics, using single and comparative case studies. The introduction forwards five recommendations for future research: (1) We need to see more comparisons across taken for granted boundaries; (2) the consequences for democracy should figure more prominently in our analysis of armed groups; (3) think more critically about standards and conceptual tools; (4) critically examine the interaction between levels of analysis; and (5) methodological pluralism would enrich the field.     

用于微机器人的线切割加工的平行弹性平台

本文给出了具有四个圆形柔性铰链的平行弹性平台的研究结果,并运用材料力学和有限元分析进行了理论仿真,利用线切割加工工艺制造了几个整体式的平台。颈缩截面的厚度为20—100um。对这些结构的刚度进行了精密测量并与理论值进行了比较。这使得能够定性的分析在电加工中产生的在微米级范围内不能忽略的粗糙度和其它表面变形的影响。通过这些实验得到的结果对其它具有几个自由度的复杂柔性机构非常有用。     

无摩擦的空气轴承技术对超精密加工机床进给驱动系统的控制需…

加工过程中随着对精密度和加工环境的要求的不断提高,越来越需要在机床的结构和控制器原理这两者之间进行特殊的组合。近年来,为了补偿进给驱动系统的直接导轨的摩擦造成的影响,人们发明了大量的控制决策,例如［１］－［３］。他们虽然提供了各式各样的功能强大的非线性的机构,但通常又使用相当复杂的计算机算法来耗费系统资源。借助于先进的空气静压轴承技术（ＦＶＭ－空气－轴承）摩擦可以被彻底消除,因而直线导轨也就不存在     

Interactive solution of bi-criteria mathematical programs

In this paper, we develop efficient interactive methods for the solution of bicriteria nonlinear programming problems. The methods do not require trade-off information from the decision maker, pose less cognitive burden and converge to the “best compromise solution” fast. Two methods, called the paired comparison method and comparative trade-off method, are presented with examples. A real application of the interactive method to a bicriteria problem that arose in the planning of the cardiovascular disease control program in the U.S. Air Force is also presented.     

Minimax and maximin facility location problems on a sphere

The problem dealt with in this article is as follows. There are n “demand points” on a sphere. Each demand point has a weight which is a positive constant. A facility must be located so that the maximum of the weighted distances (distances are the shortest arcs on the surface of the sphere) is minimized; this is called the minimax problem. Alternatively, in the maximin problem, the minimum weighted distance is maximized. A setup cost associated with each demand point may be added for generality. It is shown that any maximin problem can be reparametrized into a minimax problem. A method for finding local minimax points is described and conditions under which these are global are derived. Finally, an efficient algorithm for finding the global minimax point is constructed.