Methods for Imputing Missing Values and Synthesizing Confidential Values for Continuous and Magnitude Data

Abstract

Continuous variable is one of the major data types collected by the survey organizations. It can be incomplete such that the data collectors need to fill in the missingness. Or, it can contain sensitive information which needs protection from re-identification. One of the approaches to protect continuous microdata is to sum them up according to different cells of features. In this thesis, I represents novel methods of multiple imputation (MI) that can be applied to impute missing values and synthesize confidential values for continuous and magnitude data.

The first method is for limiting the disclosure risk of the continuous microdata whose marginal sums are fixed. The motivation for developing such a method comes from the magnitude tables of non-negative integer values in economic surveys. I present approaches based on a mixture of Poisson distributions to describe the multivariate distribution so that the marginals of the synthetic data are guaranteed to sum to the original totals. At the same time, I present methods for assessing disclosure risks in releasing such synthetic magnitude microdata. The illustration on a survey of manufacturing establishments shows that the disclosure risks are low while the information loss is acceptable.

The second method is for releasing synthetic continuous micro data by a nonstandard MI method. Traditionally, MI fits a model on the confidential values and then generates multiple synthetic datasets from this model. Its disclosure risk tends to be high, especially when the original data contain extreme values. I present a nonstandard MI approach conditioned on the protective intervals. Its basic idea is to estimate the model parameters from these intervals rather than the confidential values. The encouraging results of simple simulation studies suggest the potential of this new approach in limiting the posterior disclosure risk.

The third method is for imputing missing values in continuous and categorical variables. It is extended from a hierarchically coupled mixture model with local dependence. However, the new method separates the variables into non-focused (e.g., almost-fully-observed) and focused (e.g., missing-a-lot) ones. The sub-model structure of focused variables is more complex than that of non-focused ones. At the same time, their cluster indicators are linked together by tensor factorization and the focused continuous variables depend locally on non-focused values. The model properties suggest that moving the strongly associated non-focused variables to the side of focused ones can help to improve estimation accuracy, which is examined by several simulation studies. And this method is applied to data from the American Community Survey.

Comparison between local anaesthetic agents, lidocaine and bupivacaine, in patients undergoing third molar extraction in terms of patient satisfaction.

AIMS: (1) To determine if anaesthetic agent bupivacaine, has a prolonged effect on the period of acute postoperative pain when compared to lidocaine, a shorter acting agent. (2) To determine patient’s post-operative satisfaction and preference with regard to anaesthetic choice. METHODS: This double blind, randomised, interventional clinical trial included 85 patients. All patients had bilateral impacted lower third molars of removed under general anaesthetic. All patients received 0.5% plain bupivacaine on one randomly allocated side, with 2% lidocaine (with adrenaline) administered on the opposite side. Pain was measured using visual analogue scales at 0, 30, 60 minutes and 3, 4, 6 and 8 hours post-surgery. Pain was analysed for 1 week following surgery. Psychological evaluations and patient reported outcomes, including patient satisfaction were evaluated. RESULTS: A significant difference in pain (P=0.001) was seen during the 3-8 hour post-operative period. The upper limit of the 95% confidence interval was 10.0 or above at 3hours and 4 hours post-surgery. Two-thirds of patients preferred bupivacaine. CONCLUSION: Longer lasting anaesthetics such as bupivacaine offer a longer period of analgesia, and improve overall patient satisfaction.

ADVANCED ISLANDING DETECTION METHODS FOR SINGLE PHASE GRID CONNECTED PV INVERTERS

Due to the growing concerns associated with fossil fuels, emphasis has been placed on clean and sustainable energy generation. This has resulted in the increase in Photovoltaics (PV) units being integrated into the utility system. The integration of PV units has raised some concerns for utility power systems, including the consequences of failing to detect islanding. Numerous methods for islanding detection have been introduced in literature. They can be categorized into local methods and remote methods. The local methods are categorically divided into passive and active methods. Active methods generally have smaller Non-Detection Zone (NDZ) but the injecting disturbances will slightly degrade the power quality and reliability of the power system. Slip Mode Frequency Shift Islanding Detection Method (SMS IDM) is an active method that uses positive feedback for islanding detection. In this method, the phase angle of the converter is controlled to have a sinusoidal function of the deviation of the Point of Common Coupling (PCC) voltage frequency from the nominal grid frequency. This method has a non-detection zone which means it fails to detect islanding for specific local load conditions. If the SMS IDM employs a different function other than the sinusoidal function for drifting the phase angle of the inverter, its non-detection zone could be smaller. In addition, Advanced Slip Mode Frequency Shift Islanding Detection Method (Advanced SMS IDM), which has been introduced in this thesis, eliminates the non-detection zone of the SMS IDM. In this method the parameters of SMS IDM change based on the local load impedance value. Moreover, the stability of the system is investigated by developing the dynamical equations of the system for two operation modes; grid connected and islanded mode. It is mathematically proven that for some loading conditions the nominal frequency is an unstable point and the operation frequency slides to another stable point, while for other loading conditions the nominal frequency is the only stable point of the system upon islanding occurring. Simulation and experimental results show the accuracy of the proposed methods in detection of islanding and verify the validity of the mathematical analysis.

Local action on health inequalities: practice resources

These documents explain factors contributing to inequality in public health and set out methods for local bodies to reduce them. Documents The Marmot Review (2010) made a range of recommendations to reduce health inequalities in England. Building on the Review, the UCL Institute of Health Equity has produced 4 papers which include evidence, and examples of practical action that can be taken at a local level to reduce health inequalities. They are designed for people working in local services, particularly: directors of public health and public health teams people working in local authorities services that may influence health and wellbeing, such as planning health and wellbeing boards These practice resources build on a series of papers published in 2014 to support local action on health inequalities.

Relações interorganizacionais e eficiência coletiva: um estudo de caso no arranjo produtivo local de confecções do Agreste de Pernambuco

This work is a study in the Local Productive Arrangement of confections from Agreste of Pernambuco, as a relevant sector in economic and social aspect. This research has as central aim to understand how the inter-organizational relations influence the collective efficiency of arrangement. The theoretical framework employed highlights the approaches that deal with the benefits of business agglomeration for the develop­ment of firms and regions. It has discussed the approach of small and medium enter­ prises and industrial districts (SCHMITZ, 1997), which introduce the concept of col­ lective efficiency, explaining that only those externalities explained by Marshall (1996) are not sufficient to explain the competitive advantage of enterprises, expand­ing the idea that organizations achieve competitive advantage not acting alone. To examine the influences of relations in the collective efficiency, it has been taken as analytical perspective theory of social networks (GRANOVETTER, 1973, 1985; BURT, 1992; UZZI, 1997) because it has believe that this approach provides subsi­ dies for a structural analysis of social relationships in face the behavior of human ac­tion. By examining the organizations in a social network, you should understand the reason of this establishment of the relationship, their benefits, and as the information flow takes place and density of links between the actors (Powell; SMITH-DOERR, 1994). As for the methods, this study is characterized as a case study, in according to the purposed objectives, in addition to qualitative method. Also, due to recovering of the historical milestones of the arrangement, it is used a sectional approach with longitudinal perspective (VIEIRA, 2004). The primary and secondary data were used in order to understand the evolutionary process of the sector and their inter-actors re­ lationships in the arrangement for the promotion of development, for both, was used the contend and documentary analysis technique, respectively (DELLAGNELO ; SIL­VA, 2005). The approach of social networks has permitted understand that social re­lationships may extend the collective efficiency of the arrangement, and therefore need to develop policies that encourage the legalization of informal companies in ar­rangement, by showing up themselves representative. Thus, the relations estab­ lished in LPA of confections from Agreste of Pernambuco need for more effective mechanisms to broaden the collective efficiency. Therefore, this way as take place has directly benefited only a group of companies that are linked in some way the sup­portive institutions. So we can conclude that the inter-actor relations have limited the collective efficiency of LPA, being stimulated by the institutions in support only to groups of entrepreneurs, even those that produce external relations for all clustered companies

Alternaria Spores in the Air Across Europe: Abundance, Seasonality and Relationships with Climate, Meteorology and Local Environment

We explored the temporal and spatial variations in airborne Alternaria spore quantitative and phenological features in Europe using 23 sites with annual time series between 3 and 15 years. The study covers seven countries and four of the main biogeographical regions in Europe. The observations were obtained with Hirst-type spore traps providing time series with daily records. Site locations extend from Spain in the south to Denmark in the north and from England in the West to Poland in the East. The study is therefore the largest assessment ever carried out for Europe concerning Alternaria. Aerobiological data were investigated for temporal and spatial patterns in their start and peak season dates and their spore indices. Moreover, the effects of climate were checked using meteorological data for the same period, using a crop growth model. We found that local climate, vegetation patterns and management of landscape are governing parameters for the overall spore concentration, while the annual variations caused by weather are of secondary importance but should not be neglected. The start of the Alternaria spore season varies by several months in Europe, but the peak of the season is more synchronised in central northern Europe in the middle of the summer, while many southern sites have peak dates either earlier or later than northern Europe. The use of a crop growth model to explain the start and peak of season suggests that such methods could be useful to describe Alternaria seasonality in areas with no available observations.

Community owned solutions: identifying local best practices for social-ecological sustainability

Policies and actions that come from higher scale structures, such as international bodies and national governments, are not always compatible with the realities and perspectives of smaller scale units including indigenous communities. Yet, it is at this local social-ecological scale that mechanisms and solutions for dealing with unpredictability and change can be increasingly seen emerging from across the world. Although there is a large body of knowledge specifying the conditions necessary to promote local governance of natural resources, there is a parallel need to develop practical methods for operationalizing the evaluation of local social-ecological systems. In this paper, we report on a systemic, participatory, and visual approach for engaging local communities in an exploration of their own social-ecological system. Working with indigenous communities of the North Rupununi, Guyana, this involved using participatory video and photography within a system viability framework to enable local participants to analyze their own situation by defining indicators of successful strategies that were meaningful to them. Participatory multicriteria analysis was then used to arrive at a short list of best practice strategies. We present six best practices and show how they are intimately linked through the themes of indigenous knowledge, local governance and values, and partnerships and networks. We highlight how developing shared narratives of community owned solutions can help communities to plan governance and management of land and resource systems, while reinforcing sustainable practices by discussing and showcasing them within communities, and by engendering a sense of pride in local solutions.

Using social network analysis to identify key stakeholders in agricultural biodiversity governance and related land-use decisions at regional and local level

In 2013 the European Commission launched its new green infrastructure strategy to make another attempt to stop and possibly reverse the loss of biodiversity until 2020, by connecting habitats in the wider landscape. This means that conservation would go beyond current practices to include landscapes that are dominated by conventional agriculture, where biodiversity conservation plays a minor role at best. The green infrastructure strategy aims at bottom-up rather than top-down implementation, and suggests including local and regional stakeholders. Therefore, it is important to know which stakeholders influence land-use decisions concerning green infrastructure at the local and regional level. The research presented in this paper served to select stakeholders in preparation for a participatory scenario development process to analyze consequences of different implementation options of the European green infrastructure strategy. We used a mix of qualitative and quantitative social network analysis (SNA) methods to combine actors’ attributes, especially concerning their perceived influence, with structural and relational measures. Further, our analysis provides information on institutional backgrounds and governance settings for green infrastructure and agricultural policy. The investigation started with key informant interviews at the regional level in administrative units responsible for relevant policies and procedures such as regional planners, representatives of federal ministries, and continued at the local level with farmers and other members of the community. The analysis revealed the importance of information flows and regulations but also of social pressure, considerably influencing biodiversity governance with respect to green infrastructure and biodiversity.

Objective Methods for Characterizing Physical Exposures which may contribute to Work-related Musculoskeletal Disorders in Agricultural Workers

Thesis (Ph.D.)--University of Washington, 2016-08

Symmetric Interior Penalty DG Methods for the Compressible Navier-Stokes Equations II: Goal--Oriented A Posteriori Error Estimation

In this article we consider the application of the generalization of the symmetric version of the interior penalty discontinuous Galerkin finite element method to the numerical approximation of the compressible Navier--Stokes equations. In particular, we consider the a posteriori error analysis and adaptive mesh design for the underlying discretization method. Indeed, by employing a duality argument (weighted) Type I a posteriori bounds are derived for the estimation of the error measured in terms of general target functionals of the solution; these error estimates involve the product of the finite element residuals with local weighting terms involving the solution of a certain dual problem that must be numerically approximated. This general approach leads to the design of economical finite element meshes specifically tailored to the computation of the target functional of interest, as well as providing efficient error estimation. Numerical experiments demonstrating the performance of the proposed approach will be presented.

Comparison of different methods to detect genetic barriers in a small mammal population

Habitat fragmentation and the consequently the loss of connectivity between populations can reduce the individuals interchange and gene flow, increasing the chances of inbreeding, and the increase the risk of local extinction. Landscape genetics is providing more and better tools to identify genetic barriers.. To our knowledge, no comparison of methods in terms of consistency has been made with observed data and species with low dispersal ability. The aim of this study is to examine the consistency of the results of five methods to detect barriers to gene flow in a Mediterranean pine vole population Microtus duodecimcostatus: F-statistics estimations, Non-Bayesian clustering, Bayesian clustering, Boundary detection and Simple/Partial Mantel tests. All methods were consistent in detecting the stream as a non-genetic barrier. However, no consistency in results among the methods were found regarding the role of the highway as a genetic barrier. Fst, Bayesian clustering assignment test and Partial Mantel test identifyed the highway as a filter to individual interchange. The Mantel tests were the most sensitive method. Boundary detection method (Monmonier’s Algorithm) and Non-Bayesian approaches did not detect any genetic differentiation of the pine vole due to the highway. Based on our findings we recommend that the genetic barrier detection in low dispersal ability populations should be analyzed with multiple methods such as Mantel tests, Bayesian clustering approaches because they show more sensibility in those scenarios and with boundary detection methods by having the aim of detect drastic changes in a variable of interest between the closest individuals. Although simulation studies highlight the weaknesses and the strengths of each method and the factors that promote some results, tests with real data are needed to increase the effectiveness of genetic barrier detection.

Local savings and credit organizations : towards an innovative approach for rural Zimbabwe

Résumé : L'épargne et le crédit sont reconnus comme deux éléments clés du développement économique. Or, jusqu'à ce que les membres défavorisés d'une communauté aient accès aux ressources et services financiers, ils seront toujours privés de la participation au processus du développement et des bénéfices qui pourraient s'en suivre. La recherche indique que les services des prêts offerts par les institutions officielles ne parviennent que rarement aux plus pauvres de la société, qui sont obligés par conséquent de dépendre des intermédiaires informels comme les groupes d'épargne et les usuriers. Diverses organisations sur place comme les coopératives ont essayé de répondre aux besoins du développement des communautés défavorisées. Dans ce contexte, nous ferons d'abord le bilan historique et international des coopératives d'épargne et de crédit (i.e. les caisses populaires). Ensuite, nous analyserons quatre autres tentatives récentes qui eurent pour but de créer de nouvelles formes d'institutions financières, de les développer de telle sorte qu'elles offrent un degré d'accès raisonnable, sinon privilégié, aux ménages de revenu inférieur. L'analyse de ces cas-ci (venant du Zimbabwe, de l'Inde, du Ghana, et du Bangladesh) permettra d'identifier leurs caractéristiques communes et divergentes. À partir des résultats de cette analyse, un projet pilote au Zimbabwe fut initié pour élaborer une stratégie appropriée qui faciliterait le développement d'un réseau de caisses rurales. L'analyse théorique, la mise en pratique du projet, ainsi que les conclusions subséquentes soulignent l'importance de la participation directe des communautés à l'élaboration des organisations populaires. Il est évident que ces méthodes sont de loin plus efficaces que celles basées sur des politiques et des structures uniformes et compréhensives.||Abstract : Savings and credit are recognized as key elements of economic development, but until such time as disadvantaged members of the community have access to financial resources and services, they are obstructed from participating fully in the development process. Experience has shown that formal institutional credit bas rarely reached the poorer sectors of society, who have had to rely on informal intermediaries such as savings groups and money-lenders. Local organizations such as co-operatives have attempted to respond to the development needs of disadvantaged communities, and the historical and international record of savings and credit co-operatives (i.e. credit unions) is examined in this context. Four recent initiatives to design and develop new forms of financial institutions that give fair if not favoured access to low-income housebolds are also identified. These cases (from Zimbabwe, India, Bangladesh, and Ghana) are examined in an effort to identify common and divergent characteristics. Following from this analysis, a pilot project in Zimbabwe was initiated in an effort to elaborate an appropriate strategy for development of a network of rural savings and credit organization. The theoretical analysis, field exercise and subsequent reflections highlight the need for participatory methods of organizational design and development, rather than any all-encompassing structural or policy guidelines.

Hierarchically grouped 2D local features applied to edge contour localisation

One of the most significant research topics in computer vision is object detection. Most of the reported object detection results localise the detected object within a bounding box, but do not explicitly label the edge contours of the object. Since object contours provide a fundamental diagnostic of object shape, some researchers have initiated work on linear contour feature representations for object detection and localisation. However, linear contour feature-based localisation is highly dependent on the performance of linear contour detection within natural images, and this can be perturbed significantly by a cluttered background. In addition, the conventional approach to achieving rotation-invariant features is to rotate the feature receptive field to align with the local dominant orientation before computing the feature representation. Grid resampling after rotation adds extra computational cost and increases the total time consumption for computing the feature descriptor. Though it is not an expensive process if using current computers, it is appreciated that if each step of the implementation is faster to compute especially when the number of local features is increasing and the application is implemented on resource limited ”smart devices”, such as mobile phones, in real-time. Motivated by the above issues, a 2D object localisation system is proposed in this thesis that matches features of edge contour points, which is an alternative method that takes advantage of the shape information for object localisation. This is inspired by edge contour points comprising the basic components of shape contours. In addition, edge point detection is usually simpler to achieve than linear edge contour detection. Therefore, the proposed localization system could avoid the need for linear contour detection and reduce the pathological disruption from the image background. Moreover, since natural images usually comprise many more edge contour points than interest points (i.e. corner points), we also propose new methods to generate rotation-invariant local feature descriptors without pre-rotating the feature receptive field to improve the computational efficiency of the whole system. In detail, the 2D object localisation system is achieved by matching edge contour points features in a constrained search area based on the initial pose-estimate produced by a prior object detection process. The local feature descriptor obtains rotation invariance by making use of rotational symmetry of the hexagonal structure. Therefore, a set of local feature descriptors is proposed based on the hierarchically hexagonal grouping structure. Ultimately, the 2D object localisation system achieves a very promising performance based on matching the proposed features of edge contour points with the mean correct labelling rate of the edge contour points 0.8654 and the mean false labelling rate 0.0314 applied on the data from Amsterdam Library of Object Images (ALOI). Furthermore, the proposed descriptors are evaluated by comparing to the state-of-the-art descriptors and achieve competitive performances in terms of pose estimate with around half-pixel pose error.

A Posteriori Error Analysis of hp-Version Discontinuous Galerkin Finite Element Methods for Second-Order Quasilinear Elliptic Problems

We develop the a-posteriori error analysis of hp-version interior-penalty discontinuous Galerkin finite element methods for a class of second-order quasilinear elliptic partial differential equations. Computable upper and lower bounds on the error are derived in terms of a natural (mesh-dependent) energy norm. The bounds are explicit in the local mesh size and the local degree of the approximating polynomial. The performance of the proposed estimators within an automatic hp-adaptive refinement procedure is studied through numerical experiments.

Discontinuous Galerkin Methods for Advection-Diffusion-Reaction Problems on Anisotropically Refined Meshes

In this paper we consider the a posteriori and a priori error analysis of discontinuous Galerkin interior penalty methods for second-order partial differential equations with nonnegative characteristic form on anisotropically refined computational meshes. In particular, we discuss the question of error estimation for linear target functionals, such as the outflow flux and the local average of the solution. Based on our a posteriori error bound we design and implement the corresponding adaptive algorithm to ensure reliable and efficient control of the error in the prescribed functional to within a given tolerance. This involves exploiting both local isotropic and anisotropic mesh refinement. The theoretical results are illustrated by a series of numerical experiments.