Does what happens in group care stay in group care? The relationship between problem behaviour trajectories during care and post‐placement functioning

Residential programmes for youth may improve youth behaviour during placement, but it is not clear whether there is an association between a youth's behaviour pattern during placement and post‐placement outcomes. Life course perspective has been used to understand longitudinal patterns and pathways, and new statistical methods have been developed to identify latent trajectory groups. This study used administrative data from a family‐style group care programme to assess whether a youth's externalizing behaviour trajectory while in placement can significantly predict delinquency and adjustment outcomes at discharge and 6‐month follow‐up. Findings from multinomial logistic regression revealed a statistically significant relationship between a youth's behaviour trajectory class and outcomes. Behaviour pattern during care was a stronger predictor of outcome than cross‐sectional measures such as other demographic factors, placement history or mental‐health need indicators.     

儒学知识化与现代学术

近年来,中文学术界不时出现一种声音,批判现代儒学知识化的趋势。不少人强调:传统儒学是一套关于生命的学问,而不只是一套知识系统;它是一套整体的学问,无法被纳入现代的学科分类之中。他们忧虑:在现代西方的学术体制中,传统儒学会被割裂,因而丧失其主体性与生命力。在有关中国哲学的合法性(正当性)之争论中,也不时冒出这类的声音。这种忧虑自有其道理,但亦有其盲点。因为传统学术的专业化不仅是中国传统,也是西方传统(如西方神学)所要面对的问题。现代学术体制的形成有其历史背景,它固然有所遮蔽,亦有所开显。将传统儒学纳入现代学术体制而知识化,固然有所失,但亦有所得,因为这可以开启儒学对现代社会的批判功能。     

Causal attributions of dementia among Korean American immigrants

To better understand conceptualizations of dementia, this study explored causal attributions of dementia among 209 Korean Americans, using a self-administered questionnaire in Korean. Findings show that Korean Americans endorsed various causal attributions. Factor analysis yielded 3 dimensions of their attributions including psychological, physical/environmental, and cognitive/social. Bivariate analyses showed that younger age and higher education were related to more physical/environmental attributions, and younger age was related to more cognitive/social attributions. The study provides an understanding of causal attributions of dementia that practitioners need to understand to provide culturally competent practice and highlights a need to customize public education messages by specific ethnic groups.     

Synthetic two-way contingency tables that preserve conditional frequencies

In the area of statistical limitation, releasing synthetic data sets has become a popular method for limiting the risks of disclosure of sensitive information and at the same time maintaining analytic utility of data. However, less work has been done on how to create synthetic contingency tables that preserve some summary statistics of the original table. Studies in this area have primarily focused on generating replacement tables that preserve the margins of the original table since the latter support statistical inferences for a large set of parametric tests and models. Yet, not all synthetic tables that preserve a set of margins yield consistent results. In this paper, we propose alternative synthetic table releases. We describe how to generate complete two-way contingency tables that have the same set of observed conditional frequencies by using tools from computational algebra. We study both the disclosure risk and the data utility associated with such synthetic tabular data releases, and compare them to the traditionally released synthetic tables.     

SPARSE LOGISTIC PRINCIPAL COMPONENTS ANALYSIS FOR BINARY DATA

We develop a new principal components analysis (PCA) type dimension reduction method for binary data. Different from the standard PCA which is defined on the observed data, the proposed PCA is defined on the logit transform of the success probabilities of the binary observations. Sparsity is introduced to the principal component (PC) loading vectors for enhanced interpretability and more stable extraction of the principal components. Our sparse PCA is formulated as solving an optimization problem with a criterion function motivated from penalized Bernoulli likelihood. A Majorization-Minimization algorithm is developed to efficiently solve the optimization problem. The effectiveness of the proposed sparse logistic PCA method is illustrated by application to a single nucleotide polymorphism data set and a simulation study.     

Autoregressive model selection based on a prediction perspective

The autoregressive (AR) model is a popular method for fitting and prediction in analyzing time-dependent data, where selecting an accurate model among considered orders is a crucial issue. Two commonly used selection criteria are the Akaike information criterion and the Bayesian information criterion. However, the two criteria are known to suffer potential problems regarding overfit and underfit, respectively. Therefore, using them would perform well in some situations, but poorly in others. In this paper, we propose a new criterion in terms of the prediction perspective based on the concept of generalized degrees of freedom for AR model selection. We derive an approximately unbiased estimator of mean-squared prediction errors based on a data perturbation technique for selecting the order parameter, where the estimation uncertainty involved in a modeling procedure is considered. Some numerical experiments are performed to illustrate the superiority of the proposed method over some commonly used order selection criteria. Finally, the methodology is applied to a real data example to predict the weekly rate of return on the stock price of Taiwan Semiconductor Manufacturing Company and the results indicate that the proposed method is satisfactory.     

Cases for the nugget in modeling computer experiments

Most surrogate models for computer experiments are interpolators, and the most common interpolator is a Gaussian process (GP) that deliberately omits a small-scale (measurement) error term called the nugget. The explanation is that computer experiments are, by definition, “deterministic”, and so there is no measurement error. We think this is too narrow a focus for a computer experiment and a statistically inefficient way to model them. We show that estimating a (non-zero) nugget can lead to surrogate models with better statistical properties, such as predictive accuracy and coverage, in a variety of common situations.     

Systems Collaboration and Strategic Collaboration: Their Impacts on Supply Chain Responsiveness and Market Performance*

Recognizing the importance of interfirm collaboration and recent advancement of information technology (IT) to enhance joint decision making between firms, this study conceptualizes systems collaboration and strategic collaboration as two essential types of interfirm collaboration. The study then simultaneously examines the multiple roles of systems collaboration and strategic collaboration, and how they directly and indirectly influence a firm's supply chain responsiveness and market performance. Hypotheses are tested on survey data collected from 184 firms. Results suggest that the sequential relationships among IT competency, interfirm collaboration, and supply chain responsiveness have significant market performance implications.