Confidence intervals based on the deviance statistic for the hyperparameters in state space models

The main objective of this work is to evaluate the performance of confidence intervals, built using the deviance statistic, for the hyperparameters of state space models. The first procedure is a marginal approximation to confidence regions, based on the likelihood test, and the second one is based on the signed root deviance profile. Those methods are computationally efficient and are not affected by problems such as intervals with limits outside the parameter space, which can be the case when the focus is on the variances of the errors. The procedures are compared to the usual approaches existing in the literature, which includes the method based on the asymptotic distribution of the maximum likelihood estimator, as well as bootstrap confidence intervals. The comparison is performed via a Monte Carlo study, in order to establish empirically the advantages and disadvantages of each method. The results show that the methods based on the deviance statistic possess a better coverage rate than the asymptotic and bootstrap procedures.     

Double robust estimator in general treatment regimes based on Covariate-balancing

Double robust estimators have double the chance of being a consistent estimator of a causal effect in binary treatments cases. In this paper, we proposed an estimator of a causal effect for general treatment regimes based on covariate-balancing. Under parametrical situation, our estimator has double robustness.     

Improved two-parameter estimators for the negative binomial and Poisson regression models

Negative binomial regression (NBR) and Poisson regression (PR) applications have become very popular in the analysis of count data in recent years. However, if there is a high degree of relationship between the independent variables, the problem of multicollinearity arises in these models. We introduce new two-parameter estimators (TPEs) for the NBR and the PR models by unifying the two-parameter estimator (TPE) of Özkale and Kaç?ranlar [The restricted and unrestricted two-parameter estimators. Commun Stat Theory Methods. 2007;36:2707–2725]. These new estimators are general estimators which include maximum likelihood (ML) estimator, ridge estimator (RE), Liu estimator (LE) and contraction estimator (CE) as special cases. Furthermore, biasing parameters of these estimators are given and a Monte Carlo simulation is done to evaluate the performance of these estimators using mean square error (MSE) criterion. The benefits of the new TPEs are also illustrated in an empirical application. The results show that the new proposed TPEs for the NBR and the PR models are better than the ML estimator, the RE and the LE.     

MSE performance of the weighted average estimators consisting of shrinkage estimators when each individual regression coefficient is estimated

In this paper, we analytically derive the exact formula for the mean squared error (MSE) of two weighted average (WA) estimators for each individual regression coefficient. Further, we execute numerical evaluations to investigate small sample properties of the WA estimators, and compare the MSE performance of the WA estimators with the other shrinkage estimators and the usual OLS estimator. Our numerical results show that (1) the WA estimators have smaller MSE than the other shrinkage estimators and the OLS estimator over a wide region of parameter space; (2) the range where the relative MSE of the WA estimator is smaller than that of the OLS estimator gets narrower as the number of explanatory variables k increases.     

Towards an assessment for organizational participatory research health partnerships: A systematic mixed studies review with framework synthesis

Within the health sciences, organizational participatory research (OPR) is defined as a blend of research and action, in which academic researchers partner with health organization members. OPR is based on a sound partnership between all stakeholders to improve organizational practices. However, little research on the evaluation of OPR health partnership exists. This systematic mixed studies review sought to produce a new theoretical model that structures the evaluation of the OPR processes and related outcomes of OPR health partnerships. Six bibliographic databases were searched together with grey literature sources for OPR health partnership evaluation questionnaires. Six questionnaires were included, from which a pool of 95 OPR health partnership evaluation items were derived. The included questionnaires were appraised for the quality of their origin, development and measurement properties. A framework synthesis was performed using an existing OPR framework by organizing questionnaire items in a matrix using a hybrid thematic analysis. This led to our proposed Organizational Participatory Research Evaluation Model (OPREM) that includes three axes, Trust, Collective Learning and Sustainability (with specific dimensions) and 95 items. This model provides information to help stakeholders comprehensively structure the evaluation of their partnerships and subsequent improvement; thus, potentially helping to improve health organization practices.     

高频数据瞬时波动率核估计的窗宽选择及算法研究

在瞬时波动率的各种估计量中,非参数估计量因其能准确地度量瞬时波动率,一直是学者们的研究热点。然而,这类估计量在实际应用中都面临着最优窗宽的确定问题。由于最优窗宽中往往携带一些难以估计的未知参数,使得在实际应用过程中确定最优窗宽的具体数值存在困难。本文以瞬时波动率的核估计量为例,借鉴非参数回归分析中窗宽选择的思想,构建了一种能从数据中准确计算出最优窗宽具体值的算法。理论的分析和数值上的验证表明：文中所构建的算法具有良好的稳定性、适应性和收敛速度。算法的提出为瞬时波动率的后续应用研究铺平道路。     

基于股价预测的泛证券投资组合策略

本文基于期望效用最大化和L₁-中位数估计研究了在线投资组合选择问题。与EG（Exponential Gradient）策略仅利用单期价格信息估计价格趋势不同,本文将利用多期价格信息估计价格趋势,以提高在线策略的性能。首先,基于多期价格数据,利用L₁-中位数估计得到预期价格趋势。然后,通过期望效用最大化,提出一个新的具有线型时间复杂度的在线策略,EGLM（Exponential Gradient via L₁-Median）。并通过相对熵函数定义资产权重向量的距离,进而证明了EGLM策略具有泛证券投资组合性质。最后,利用国内外6个证券市场的历史数据进行实证分析,结果表明相较于UP（Universal Portfolio）策略和EG策略,EGLM策略有更好的竞争性能。     

Measuring the health patterns of the ‘mixed/multiple’ ethnic group in Britain: data quality problems,reporting issues,and implications for policy

The ‘mixed’ group, officially recognised in the 2001 Census, is one of the most rapidly growing ethnic groups in Britain. Although ‘mixed’ categorisation was added to ethnic coding in NHS datasets, our knowledge of health patterns for this population is meagre. Data quality problems remain a key obstacle, including poor reproducibility of the data and constraints on reporting due to sparse data bias. The consequent minimal and indicative evidence base has focused mainly on risky health behaviours, mental health and generic measures of self-rated health, as it has in the U.S.A. and Canada. There is negligible information on the main underlying causes of death, such as neoplasms, heart disease and stroke. Consideration should be given to pooling data across multiple years of health and general purpose surveys to enable reporting for the four ‘mixed’ categories and adjustment for mediating factors and relevant confounders, such as measures of socio-economic status.     

Estimating the effects of network covariates on subgroup insularity with a hierarchical mixed membership stochastic blockmodel

Social networks describe the relationships and interactions among a group of individuals. In many peer relationships, individuals tend to associate more often with some members than others, forming subgroups or clusters. Subgroup structure varies across networks; subgroups may be insular, appearing distinct and isolated from one another, or subgroups may be so integrated that subgroup structure is not visually apparent, and there are numerous ways of quantifying these types of structures. We propose a new model that relates the amount of subgroup integration to network attributes, building on the mixed membership stochastic blockmodel (Airoldi et al., 2008) and subsequent work by Sweet and Zheng (2017) and Sweet et al. (2014). We explore some of the operating characteristics of this model with simulated data and apply this model to determine the relationship between teachers’ instructional practices and their classrooms’ peer network subgroup structure.     

Developing a methodology to assess the impact of research grant funding: A mixed methods approach

This paper discusses the development of a mixed methods approach to analyse research funding. Research policy has taken on an increasingly prominent role in the broader political scene, where research is seen as a critical factor in maintaining and improving growth, welfare and international competitiveness. This has motivated growing emphasis on the impacts of science funding, and how funding can best be designed to promote socio-economic progress. Meeting these demands for impact assessment involves a number of complex issues that are difficult to fully address in a single study or in the design of a single methodology. However, they point to some general principles that can be explored in methodological design. We draw on a recent evaluation of the impacts of research grant funding, discussing both key issues in developing a methodology for the analysis and subsequent results. The case of research grant funding, involving a complex mix of direct and intermediate effects that contribute to the overall impact of funding on research performance, illustrates the value of a mixed methods approach to provide a more robust and complete analysis of policy impacts. Reflections on the strengths and weaknesses of the methodology are used to examine refinements for future work.