Importance sampling type estimators based on approximate marginal Markov chain Monte Carlo

We consider importance sampling (IS) type weighted estimators based on Markov chain Monte Carlo (MCMC) targeting an approximate marginal of the target distribution. In the context of Bayesian latent variable models, the MCMC typically operates on the hyperparameters, and the subsequent weighting may be based on IS or sequential Monte Carlo (SMC), but allows for multilevel techniques as well. The IS approach provides a natural alternative to delayed acceptance (DA) pseudo-marginal/particle MCMC, and has many advantages over DA, including a straightforward parallelization and additional flexibility in MCMC implementation. We detail minimal conditions which ensure strong consistency of the suggested estimators, and provide central limit theorems with expressions for asymptotic variances. We demonstrate how our method can make use of SMC in the state space models context, using Laplace approximations and time-discretized diffusions. Our experimental results are promising and show that the IS-type approach can provide substantial gains relative to an analogous DA scheme, and is often competitive even without parallelization.     

Optimal asset allocation for participating contracts with mortality risk under minimum guarantee

Abstract

We investigate an optimal investment problem of participating insurance contracts with mortality risk under minimum guarantee. The insurer aims to maximize the expected utility of the terminal payoff. Due to its piecewise payoff structure, this optimization problem is a non-concave utility maximization problem. We adopt a concavification technique and a Lagrange dual method to solve the problem and derive the representations of the optimal wealth process and trading strategies. We also carry out some numerical analysis to show how the portfolio insurance constraint impacts the optimal terminal wealth.     

基于犹豫直觉模糊语言集均值-标准差偏好距离的多属性决策研究

在犹豫直觉模糊语言集和语言尺度函数的基础上定义了均值-标准差偏好的Hamming距离，并提出了基于犹豫直觉模糊语言集距离TOPSIS和TODIM的多属性决策方法，进一步利用这两种方法对实例问题建筑商的招标方案进行排序，并讨论了偏好参数对排序结果的灵敏度分析。为验证上述方法的有效性，与已有方法对同一实例的排序结果进行了比较分析。     

Evaluating and Visualizing the Economic Impact of Commercial Districts Due to an Electric Power Network Disruption

Critical infrastructure networks enable social behavior, economic productivity, and the way of life of communities. Disruptions to these cyber–physical–social networks highlight their importance. Recent disruptions caused by natural phenomena, including Hurricanes Harvey and Irma in 2017, have particularly demonstrated the importance of functioning electric power networks. Assessing the economic impact (EI) of electricity outages after a service disruption is a challenging task, particularly when interruption costs vary by the type of electric power use (e.g., residential, commercial, industrial). In contrast with most of the literature, this work proposes an approach to spatially evaluate EIs of disruptions to particular components of the electric power network, thus enabling resilience‐based preparedness planning from economic and community perspectives. Our contribution is a mix‐method approach that combines EI evaluation, component importance analysis, and GIS visualization for decision making. We integrate geographic information systems and an economic evaluation of sporadic electric power outages to provide a tool to assist with prioritizing restoration of power in commercial areas that have the largest impact. By making use of public data describing commercial market value, gross domestic product, and electric area distribution, this article proposes a method to evaluate the EI experienced by commercial districts. A geospatial visualization is presented to observe and compare the areas that are more vulnerable in terms of EI based on the areas covered by each distribution substation. Additionally, a heat map is developed to observe the behavior of disrupted substations to determine the important component exhibiting the highest EI. The proposed resilience analytics approach is applied to analyze outages of substations in the boroughs of New York City.     

Explicit analytical solutions for ARL of CUSUM chart for a long-memory SARFIMA model

This paper aims to derive explicit analytical solutions for Average Run Length (ARL) of CUSUM chart for the SARFIMA(P,D,Q)_S process with exponential white noise. Measurement of performance was done with the ARL in terms of percentage error and CPU time. The results obtained from the explicit formulas were compared focusing on the performance using the numerical integral equation (NIE) method. Both methods had similarly excellent agreement with the percentage error at less than 0.25%. Meanwhile, the explicit formulas consumed less CPU time than the NIE method. It is clear that the explicit formulas are a good alternative in real applications.     

Interview schedule development for a Sequential explanatory mixed method design: complementary-alternative medicine (CAM) study among Indonesian psychologists

Sequential explanatory mixed method design is the most frequently applied in both health and social sciences literature. It is denoted by ‘QUAN → qual’ which represents the quantitative study occurs first and has greater weight in addressing the study’s aims, and the qualitative study follows to explain quantitative results. Despite the extensive use of sequential explanatory design, there are limited references to this design. Therefore, this methodological paper attempts to fill the gap by providing an illustration in developing a sequential explanatory interview schedule based on complementary-alternative medicine (CAM) study among clinical psychologists in Indonesia. The most important step to develop sequential explanatory interview schedule was the construction of aspects and questions that were immensely grounded on the most notable quantitative results. In this study, eight aspects of interview schedule were constructed after analyses of the nationwide survey. The interview schedule then piloted among participants with fairly similar characteristics to the participants in the main interviews. This process enhanced the quality of questions through feedback from participants and improved the interviewer’s skills through familiarization with questions. It is expected that this reflection report could be adopted as a practical guideline in developing interview schedule for sequential explanatory mixed method design, particularly in the field of psychology.     

植入式教学法在大学生职业生涯规划课中的应用

大学生职业生涯规划课是一门集理论型、实践型、经验型为一体的综合性课程。在课堂教学中，通过背景知识、色彩图案、哲理案例、互动体验的植入，充实优化教学内容，增强课程吸引力，激发学生的学习兴趣。通过探索启迪学生思想、更新学习观念以及差异化职业能力培养的教学方法，满足学生的个性化职业规划学习需求。     

FDA: strong consistency of the kNN local linear estimation of the functional conditional density and mode

In this paper we present a new estimator of the conditional density and mode when the co-variables are of functional kind. This estimator is a combination of both, the k-Nearest Neighbours procedure and the functional local linear estimation. Then, for each statistical parameter (conditional density or mode), results concerning the strong consistency and rate of convergence of the estimators are presented. Finally, their performances, for finite sample sizes, are illustrated by using simulated data.     

Carbon Emissions Abatement (CEA) allocation and compensation schemes based on DEA

As environment constraints on economic growth are strengthening, Carbon Emissions Abatement (CEA) allocation becomes a significant issue that draws academia׳s attention. In the literature, the Data Envelopment Analysis (DEA) technique has been applied to obtain CEA allocation with centralized models. Nevertheless, a centralized allocation plan suffers from an implementation difficulty in persuading decision-making units (DMUs) into an agreement. In this paper, we propose a new two-step method to mitigate this side effect. In the first step, we provide improved DEA-based centralized allocation models under the assumptions of constant returns-to-scale (CRS) and variable returns-to-scale (VRS) respectively and in the second step, two compensation schemes are developed for centralized allocation plans. An empirical application to the countries in Organization for Economic Co-operation and Development (OECD) is presented to elaborate the main idea.     

HDDA: DataSifter: statistical obfuscation of electronic health records and other sensitive datasets

There are no practical and effective mechanisms to share high-dimensional data including sensitive information in various fields like health financial intelligence or socioeconomics without compromising either the utility of the data or exposing private personal or secure organizational information. Excessive scrambling or encoding of the information makes it less useful for modelling or analytical processing. Insufficient preprocessing may compromise sensitive information and introduce a substantial risk for re-identification of individuals by various stratification techniques. To address this problem, we developed a novel statistical obfuscation method (DataSifter) for on-the-fly de-identification of structured and unstructured sensitive high-dimensional data such as clinical data from electronic health records (EHR). DataSifter provides complete administrative control over the balance between risk of data re-identification and preservation of the data information. Simulation results suggest that DataSifter can provide privacy protection while maintaining data utility for different types of outcomes of interest. The application of DataSifter on a large autism dataset provides a realistic demonstration of its promise practical applications.