Sensitivity Analysis Using Risk Measures

In a quantitative model with uncertain inputs, the uncertainty of the output can be summarized by a risk measure. We propose a sensitivity analysis method based on derivatives of the output risk measure, in the direction of model inputs. This produces a global sensitivity measure, explicitly linking sensitivity and uncertainty analyses. We focus on the case of distortion risk measures, defined as weighted averages of output percentiles, and prove a representation of the sensitivity measure that can be evaluated on a Monte Carlo sample, as a weighted average of gradients over the input space. When the analytical model is unknown or hard to work with, nonparametric techniques are used for gradient estimation. This process is demonstrated through the example of a nonlinear insurance loss model. Furthermore, the proposed framework is extended in order to measure sensitivity to constant model parameters, uncertain statistical parameters, and random factors driving dependence between model inputs.     

Multi-ethnic school environment from the school leader’s perspective: challenges and approaches to improve multi-cultural competency among teachers in Malaysia

This study aims to identify the challenges and approaches to improve a multicultural competency among teachers in multi-ethnic schools environment based on school leaders’ perspective. Data collection method used in this study is qualitative method. A total of nine school leaders from the states of Kedah, Penang, and Perak were included in the interview session. The results of this study show that the challenges faced by teachers in practicing the multicultural competency are lack of experience toward cultural diversity. A late exposure to cultural diversity makes it difficult for teachers to accept and adapt in a multi-ethnic school environment. The approaches suggested by school leaders to improve the multicultural competency are as follows: (1) Courses on cultural diversity should be organized to enhance the knowledge about cultural diversity; (2) the development of more open and better policy for education system in Malaysia in order to prepare teachers with a higher multicultural competency for teaching students from diverse ethnic and cultural groups; (3) an early exposure to cultural diversity in order to increase the multicultural competency, especially during teaching practicum.     

Preschoolers' Attachment to Mother and Risk for Adjustment Problems in Kindergarten: Can Teachers Make a Difference?

Based on attachment theory, two aims were addressed. Firstly, we tested whether close teacher–child relationships may buffer children who are less securely attached to their mothers against negative outcomes, such as aggressive behavior. Secondly, our study evaluated whether teacher sensitivity may protect less securely attached children against forming less close relationships with their teachers. In a sample of 127 children, mother–child attachment was observed in preschool. In kindergarten, teacher sensitivity was observed, and teacher–child closeness and child aggressive behavior were rated by the teacher. Results of multilevel hierarchical regression analyses first showed that with high teacher–child closeness, less securely attached children are no longer at risk for more aggressive behavior. Secondly, it was found that with high teacher sensitivity, less securely attached children are no longer at risk for developing less close relationships with their teachers.     

董事会非正式层级与高管薪酬契约有效性

董事会治理是影响高管薪酬契约有效性的关键,已有研究主要从董事会结构层面展开,缺乏对董事会内部运作机制与高管薪酬契约有效性关系的具体分析。基于董事权威不平衡性的治理效应视角,系统考察了董事会非正式层级与高管薪酬契约有效性的逻辑关系及其作用机理,得出了一些具有重要价值的结论,主要包括:1.董事会非正式层级对高管薪酬业绩敏感性具有显著的负向调节作用,且主要体现在业绩下滑的样本公司中,表明董事会非正式层级对高管薪酬契约的影响主要体现在公平性和风险匹配性方面。2.只有当最高层级董事身份为独立董事时,董事会非正式层级对高管薪酬业绩敏感性具有显著的负向调节效应。3.董事会非正式层级增强了高管薪酬的行业和地区薪酬敏感性,进一步验证了董事会非正式层级的社会公平效应。论文的研究为从微观运作层面理解董事会治理和高管薪酬契约有效性提供一种新的视角和结论,对于优化董事会治理和提高高管薪酬契约有效性等均具有重要借鉴意义。     

感悟化：《东方的太阳》拓展政治抒情诗的新路径

谭仲池先生的《东方的太阳》,把抒情建立在与客体的心灵对话之上,结构穿梭在“求谒”的且歌且行之中,语言紧贴智性化的生命律动,作品以感悟的方式丰富、拓展了政治抒情诗的新路径。     

Symmetrical Design of Experiment in Global Sensitivity Analysis Based on ANOVA High-dimensional Model Representation

This article reviews symmetrical global sensitivity analysis based on the analysis of variance of high-dimensional model representation. To overcome the computational difficulties and explore the use of symmetrical design of experiment (SDOE), two methods are presented. If the form of the objective function f is known, we use SDOE to estimate the symmetrical global sensitivity indices instead of Monte Carlo or quasi-Monte Carlo simulation. Otherwise, we use the observed values of the experiment to do symmetrical global sensitivity analysis. These methods are easy to implement and can reduce the computational cost. An example is given by symmetrical design of experiment.     

Local influence diagnostics for incomplete overdispersed longitudinal counts

We develop local influence diagnostics to detect influential subjects when generalized linear mixed models are fitted to incomplete longitudinal overdispersed count data. The focus is on the influence stemming from the dropout model specification. In particular, the effect of small perturbations around an MAR specification are examined. The method is applied to data from a longitudinal clinical trial in epileptic patients. The effect on models allowing for overdispersion is contrasted with that on models that do not.     

Local predictive influence in bayesian linear models with conjugate priors

Cook (1986) presented the idea of local influence to study the sensitivity of inferences to model assumptions:introduce a vector δ of perturbations to the model; choose a discrepancy function D to measure differences between the original inference and the inference under the perturbed model; study the behavior of D near δ = 0, the original model, usually by taking derivatives. Johnson and Geisser (1983) measure influence in Bayesian inference by the Kullback-Leibler divergence between predictive distributions. I~IcCulloch (1989) is a synthesis of Cook and Johnson and Geisser, using Kullback-Leibler divergence between posterior or predictive distributions as the discrepancy function in Bayesian local influence analyses. We analyze a special case for which McCulloch gives the general theory; namely, the linear model with conjugate prior. We present specific formulae for local influence measures for 1) changes in the parameters of the gamma prior for the precision, 2) changes in the mean of the normal prior for the regression coefficients, 3) changes in the covariance matrix of the normal prior for the regression coefficients and 4) changes in the case weights. Our method is an easy way to find locally influential subsets of points without knowing in advance the sizes of the subsets. The techniques are illustrated with a regression example.     

Reliability of global sensitivity indices

Uncertainty and sensitivity analysis is an essential ingredient of model development and applications. For many uncertainty and sensitivity analysis techniques, sensitivity indices are calculated based on a relatively large sample to measure the importance of parameters in their contributions to uncertainties in model outputs. To statistically compare their importance, it is necessary that uncertainty and sensitivity analysis techniques provide standard errors of estimated sensitivity indices. In this paper, a delta method is used to analytically approximate standard errors of estimated sensitivity indices for a popular sensitivity analysis method, the Fourier amplitude sensitivity test (FAST). Standard errors estimated based on the delta method were compared with those estimated based on 20 sample replicates. We found that the delta method can provide a good approximation for the standard errors of both first-order and higher-order sensitivity indices. Finally, based on the standard error approximation, we also proposed a method to determine a minimum sample size to achieve the desired estimation precision for a specified sensitivity index. The standard error estimation method presented in this paper can make the FAST analysis computationally much more efficient for complex models.     

Relevance of Risk Predictions Derived from a Chronic Species Sensitivity Distribution with Cadmium to Aquatic Populations and Ecosystems

Criteria to protect aquatic life are intended to protect diverse ecosystems, but in practice are usually developed from compilations of single‐species toxicity tests using standard test organisms that were tested in laboratory environments. Species sensitivity distributions (SSDs) developed from these compilations are extrapolated to set aquatic ecosystem criteria. The protectiveness of the approach was critically reviewed with a chronic SSD for cadmium comprising 27 species within 21 genera. Within the data set, one genus had lower cadmium effects concentrations than the SSD fifth percentile‐based criterion, so in theory this genus, the amphipod Hyalella, could be lost or at least allowed some level of harm by this criteria approach. However, population matrix modeling projected only slightly increased extinction risks for a temperate Hyalella population under scenarios similar to the SSD fifth percentile criterion. The criterion value was further compared to cadmium effects concentrations in ecosystem experiments and field studies. Generally, few adverse effects were inferred from ecosystem experiments at concentrations less than the SSD fifth percentile criterion. Exceptions were behavioral impairments in simplified food web studies. No adverse effects were apparent in field studies under conditions that seldom exceeded the criterion. At concentrations greater than the SSD fifth percentile, the magnitudes of adverse effects in the field studies were roughly proportional to the laboratory‐based fraction of species with adverse effects in the SSD. Overall, the modeling and field validation comparisons of the chronic criterion values generally supported the relevance and protectiveness of the SSD fifth percentile approach with cadmium.