Measuring the Unmeasurable—A Study of Domestic Violence Risk Prediction and Management

ABSTRACT

The prevention of domestic violence (DV) have aroused serious concerns in Taiwan because of the disparity between the increasing amount of reported DV cases that doubled over the past decade and the scarcity of social workers. However, most common collaborations for DV prevention, between academic researchers and advocacy groups or governments outsourcing, often fail to produce effective prevention strategies. Hence, Data for Social Good Initiative (D4SG) worked with Taipei City Government to improve the efficiency of DV prevention and risk management on two levels—project collaboration level and data level. On the project collaboration level, we adopted a platform strategy and utilize public-private partnership (PPP) to connect and empower change agents across data silos from pilot runs to actual project execution. On the data level, we helped social workers differentiate the risk level of new cases by building a repeat victimization risk prediction model using random forest method with the 2015 data from Taipei City government’s DV database. The accuracy and F1-measure of our model were 96.3% and 62.8%. This projects’ PPP approach and quantification method successfully improved DV prevention process. These methodologies have also been applied to other work fields including firework prediction, emergency healthcare management as a paradigm.     

Food for Thought: Analyzing Public Opinion on the Supplemental Nutrition Assistance Program

ABSTRACT

This project explores public opinion on the Supplemental Nutrition Assistance Program (SNAP) in news and social media outlets, and tracks elected representatives’ voting records on issues relating to SNAP and food insecurity. We used machine learning, sentiment analysis, and text mining to analyze national and state level coverage of SNAP in order to gauge perceptions of the program over time across these outlets. Results indicate that the majority of news coverage has negative sentiment, more partisan news outlets have more extreme sentiment, and that clustering of negative reporting on SNAP occurs in the Midwest. Our final results and tools will be displayed in an online application that the ACFB Advocacy team can use to inform their communication to relevant stakeholders.     

Factors influencing enrolment: A case study from Birth to Twenty, the 1990 birth cohort in Soweto–Johannesburg

Longitudinal studies offer significant advantages in rendering data commensurate with the complexity of human development. However, incomplete enrolment and attrition over time can introduce bias. Furthermore, there is a scarcity of evaluative information on cohorts in developing countries. This paper documents various strategies adopted to minimize loss to follow up and describes a retrospective analysis of a small group of families who were missed during initial enrolment and through several subsequent rounds of data collection of the Birth to Twenty (BT20) birth cohort in Soweto–Johannesburg, South Africa that began in 1990. A purposive case study approach was used, and 10 of the 119 families missed at enrolment were interviewed to investigate why these families were not enrolled into the study. The findings demonstrate that high mobility, both within urban areas and between urban and rural areas, are a major challenge for longitudinal studies in densely populated urban areas. In addition, enrolment was also affected by individuals changing their names, largely motivated to facilitate access to employment under Apartheid, as well as varying motivations for participating in research. Longitudinal studies in the developing country context must be mindful of the political, social and economic climate that influences enrolment and ongoing cohort maintenance.     

Automatic and asymptotically optimal data sharpening for nonparametric regression

In this article we consider data-sharpening methods for nonparametric regression. In particular modifications are made to existing methods in the following two directions. First, we introduce a new tuning parameter to control the extent to which the data are to be sharpened, so that the amount of sharpening is adaptive and can be tuned to best suit the data at hand. We call this new parameter the sharpening parameter. Second, we develop automatic methods for jointly choosing the value of this sharpening parameter as well as the values of other required smoothing parameters. These automatic parameter selection methods are shown to be asymptotically optimal in a well defined sense. Numerical experiments were also conducted to evaluate their finite-sample performances. To the best of our knowledge, there is no bandwidth selection method developed in the literature for sharpened nonparametric regression.     

Consistent model selection and data-driven smooth tests for longitudinal data in the estimating equations approach

Summary.  Model selection for marginal regression analysis of longitudinal data is challenging owing to the presence of correlation and the difficulty of specifying the full likelihood, particularly for correlated categorical data. The paper introduces a novel Bayesian information criterion type model selection procedure based on the quadratic inference function, which does not require the full likelihood or quasi-likelihood. With probability approaching 1, the criterion selects the most parsimonious correct model. Although a working correlation matrix is assumed, there is no need to estimate the nuisance parameters in the working correlation matrix; moreover, the model selection procedure is robust against the misspecification of the working correlation matrix. The criterion proposed can also be used to construct a data-driven Neyman smooth test for checking the goodness of fit of a postulated model. This test is especially useful and often yields much higher power in situations where the classical directional test behaves poorly. The finite sample performance of the model selection and model checking procedures is demonstrated through Monte Carlo studies and analysis of a clinical trial data set.     

数据挖掘方法应用于调查数据的抽样权重问题

 在将数据挖掘方法应用于抽样调查数据时,会遇到抽样权重的处理问题。本文提出采用放回的、与样本单元权数大小成比例的再抽样方法,简称PPWWR再抽样,来实现“事后”自加权设计。实现“事后”自加权设计后的子样本可忽略掉样本权数,直接采用常规的图示方法和数据挖掘算法进行分析。随后,基于2007中国公民科学素质调查贵州省数据,通过模拟分析讨论了PPWWR再抽样子样本的样本量问题,发现 是一个比较合适的样本量。这一结论可能为其它大型复杂抽样调查数据的数据挖掘实施问题提供借鉴。     

对统计调查质量特性的探讨

 在界定统计调查质量含义的基础上,总结了文献中统计调查质量维度。按照统计调查过程和统计调查主体纵横两个方向分别对统计质量特性进行分解,并对纵横两个方向进行整合分析。最后给出这种分解的统计质量评估、管理和改进的应用。     

A Composite Indicator for University Quality Assesment: The Case of Spanish Higher Education System

This study presents a synthetic indicator for quality assessment in the Spanish public university system. The indicator is based on a series of components and simple indicators that were obtained from the public universities’ financial planning estimates. The paper takes a quantitative, wide-ranging approach to analyse the quality of university institutions and is intended to be complimentary to other qualitative approaches. Data Envelopment Analysis was used in order to facilitate the aggregation and weighting of the data used to construct the synthetic indicator. This technique allows the analyst to endogenously determine the weighting of the partial indicators while respecting the peculiarities intrinsic to each university. The results reveal that there are significant differences among Spanish public universities. In addition to how relatively well each of the institutions performs, other factors would seem to be influencing the results; essentially these are related to socio-economic factors and to the application of university policy in Spain over recent decades.

封闭型基金绩效评估与相对投资价值评价

本文首先讨论了封闭型基金绩效评估与投资价值评价的关系,提出了基于DEA的非参数型基金相对投资价值评价模型和研判准则,并以我国基金市场为例,对比研究了DEA相对价值指数与Sharpe绩效指数的投资指导意义。