决策可靠性分析及在SAR图像目标识别中的应用

决策融合是提高合成孔径雷达（Synthetic Aperture Radar,SAR）目标识别性能的重要手段，然而，可靠性较弱的决策往往会导致最终决策融合的效果变差。将可靠性分析引入基于决策融合的SAR目标识别方法中，分别计算各个决策的可靠性系数并选取可靠性的决策参与最终的决策融合。为了验证方法的有效性，分别将提出的可靠性分析应用于多特征决策融合以及多分类器决策融合并基于MSTAR（Moving and Stationary Target Acquisition and Recognition）数据集进行了目标识别实验。在基于主成分分析、线性鉴别分析和非负矩阵分解三种特征进行多特征决策融合的条件下，所提方法和直接进行决策融合的方法的识别率分别为97.47%和96.50%。在基于K近邻、支持向量机和稀疏表示分类器的多分类器决策融合中，所提方法和直接进行决策融合的方法的识别率分别为97.10%和96.28%。实验结果充分证明了所提方法的有效性。     

A Statistical Model to Investigate the Reproducibility Rate Based on Replication Experiments

The reproducibility crisis, that is, the fact that many scientific results are difficult to replicate, pointing to their unreliability or falsehood, is a hot topic in the recent scientific literature, and statistical methodologies, testing procedures and p‐values, in particular, are at the centre of the debate. Assessment of the extent of the problem–the reproducibility rate or the false discovery rate–and the role of contributing factors are still an open problem. Replication experiments, that is, systematic replications of existing results, may offer relevant information on these issues. We propose a statistical model to deal with such information, in particular to estimate the reproducibility rate and the effect of some study characteristics on its reliability. We analyse data from a recent replication experiment in psychology finding a reproducibility rate broadly coherent with other assessments from the same experiment. Our results also confirm the expected role of some contributing factor (unexpectedness of the result and room for bias) while they suggest that the similarity between original study and the replica is not so relevant, thus mitigating some criticism directed to replication experiments.     

Covariance breakdowns and connectedness of crude oil futures markets with non-synchronous data

ABSTRACT

This article identifies the breakdowns in the covariance of three benchmark crude oil futures markets (WTI, Brent and Dubai) and investigates the changes of market connectedness across the breakdown periods. As the crude oil futures are traded in different regions, this article eliminates the non-synchronous trading data by employing the Vector Moving Average structure and the Bayesian data augmentation approach, which keeps the integrity of original data without changing its properties. The results show that there are significant breaks in the covariance structure of crude oil futures markets. The breakdown periods are consistent with the periods when the market volatilities are at high level and the returns are volatile. The changes of market connectedness are independent of the covariance states, which supports the globalization hypothesis for the crude oil market. The results also suggest that there is more information flow out of the WTI than to the WTI during the sample period, particularly during the breakdown periods in 2008–2009.     

Stress and turnover intents in international organizations: social support and work–life balance as resources

Abstract

This study investigates whether work opportunities have an impact on stress and the related turnover intentions of employees working in intergovernmental international organizations (IOs). It contextualizes the job resources and demands model within IOs’ specific work conditions. The empirical test is based on original data from a survey administered in four major organizations of the United Nations system. Results demonstrate that social work opportunities and work–life balance are organizational levers reducing stress and willingness to quit for employees who are facing red tape or the stresses of being an expatriate. In this context, the relationships between these work opportunities and turnover intention are partially mediated by stress. Contextualized HR management propositions are made to help organizations coping with these management challenges.     

Emotions,deliberations, and end‐of‐life products

Death is inevitable; yet, not all consumers prepare for death by purchasing end‐of‐life (EOL) products. Using the theory of reasoned action (TRA) and the dual‐process model framework, this study aims to examine the role of emotions and cognitions in influencing consumers' decisions to engage in planning for death. A mixed methodology design was used. Study 1, a qualitative study, uncovered positive and negative emotions and deliberative reasoning that comprise consumers' EOL purchase decision process. Study 2, a quantitative study, confirmed that emotions and deliberations independently and jointly influenced consumers' EOL attitude and behavior and that emotions affected deliberations for both prepaid funerals and wills. Subjective norms outperformed attitude in predicting both products' purchase behavior. These finding supported the dual‐process model of behavior and the TRA in the EOL research context and contributed to the EOL literature by investigating the effects of emotions and deliberations concurrently; thus validating the important role of emotions in influencing EOL planning and purchase. In light of our findings, marketers could, after due cognizance of the morbidity and sensitivity of the topic, develop actionable promotional and segmentation strategies for EOL products and other emotion‐laden, unsought products and service.     

A spatial autoregressive stochastic frontier model for panel data incorporating a model of technical inefficiency

By integrating Battese and Coelli’s (1995) model and the spatial autoregressive model (SAR), a spatial autoregressive stochastic frontier model for panel data is developed. The main feature of this frontier model is a spatial lag term of explained variables and the joint structure of a production possibility frontier with a model of technical inefficiency. The model addresses both spatial dependence and heteroskedastic technical inefficiency. This study applies maximum likelihood methods considering the endogenous spatial lag term. The proposed model nests several existing models. Further, an empirical analysis using data on the Japanese manufacturing industry is conducted and the existing models are tested against the proposed model, which is found to be statistically supported. The findings suggest that estimates in the existing spatial and non-spatial models may exhibit bias because of lack of determinants of technical inefficiency, as well as a spatial lag. This bias also affects the technical efficiency score and its ranking.     

Club convergence and drivers of digitalization across Indian states

Disparity in the level of digitalization is a crucial driver of economic inequality in an economy. Although a pocket of its population is still bereft of the benefits of digitalization, India currently has the second highest number of internet subscribers in the world despite the nation's late adoption of digital technology. An accurate assessment of the current state of digitalization in the country is required for devising effective initiatives towards building a Digital India and bridging the nation's internal digital divide. Considering 17 major Indian states and 21 variables for 10 years, we constructed a composite index of digitalization with the help of Principal Component Method (PCM). This paper identifies factors responsible for the digitalization divide across states. In the second part of the analysis, the study confirms, with the help of club convergence test, the absence of overall convergence towards digitalization and the existence of the non-convergent group. This paper also shows that the non-convergent group of states are at the bottom of the ranking table which indicates the need for greater attention to initiatives to bridge the digital divide.     

Corporate social responsibility and competitive advantage: A developing country perspective

Corporate social responsibility (CSR) has attracted wider research interests over recent decades. While some studies have examined the impact of CSR activities on firm competitive advantage (CA), the findings so far remain contradictory. Moreover, the role of export orientation, firm strategy, and structure on the association between CSR and CA has not been explicitly examined. Thus, the purpose of this study is to examine the moderating role of export orientation, firm strategy, structure, and firm size on the association between CSR and CA. Using a sample of 179 responses from management staff in organizations across five sectors in a developing country context of Ghana, the study found positive effects of CSR on CA. The study contributes to the resource‐based view (RBV) scholarship by confirming the important complementary effect of export orientation and organizational structure as important resources and capabilities on the CSR–competitiveness relationship. However, no evidence of a moderating effect of firm strategy, or firm size on the CSR–CA relationship was found. These findings are instructive, impactful, and enrich the existing literature on CSR and strategy. Implications for theory and practice are also discussed.     

The Non‐Observed Economy in Portugal: The monetary model and the MIMIC model

This paper offers estimations for the Portuguese path of the Non‐Observed Economy (NOE), in the period 1970–2015, through two seminal approaches: monetary method and the Multiple Indicators Multiple Causes (MIMIC) model. It is observed that the tax burden and social benefits are its main causes. Then, to get a more in‐depth understanding of the phenomenon, it provides a study of the Granger causality between the NOE and the official Gross Domestic Product (GDP), emphasizing the implications of the NOE on the Portuguese economy. Evidence has been found for the existence of bidirectional causality between the NOE and the GDP, suggesting that the formal economy affects the NOE, and conversely that the NOE affects the economic growth.     

Failure to learn from failure: The 2008 mortgage crisis as a déjà vu of the mortgage meltdown of 1994

This article traces the developments in the market for residential mortgage-backed securities (MBS) during the period 1970–2008. Drawing on an analysis of trade publications, business press, and interviews with practitioners, it shows that an MBS market meltdown in 1994 provided clear signals of problems with MBS. The market participants did not re-evaluate their use of risk management tools or adjust security design in response to the 1994 crisis, suggesting a lack of understanding of the implications of the crisis. The 1994 meltdown showed that MBS were vulnerable to systematic risks and that these risks could precipitate an MBS market crash. Furthermore, the 1994 meltdown demonstrated that large-scale investment in MBS could affect the primary mortgage market, thereby rendering the MBS risks unpredictable. After 1994, MBS investment shifted to MBS backed by mortgages with default risk – a development that led to the crash of 2008. By drawing parallels between the 1994 and 2008 crises, this article shows how the MBS market failed to self-correct. The results suggest that financial market participants do not always incorporate relevant information in their decision-making and that market participants have difficulties in both foreseeing the effect of financial innovations on markets and interpreting these effects.