Inferential Implications of Over-Parametrization: A Case Study in Incomplete Categorical Data

P>In the context of either Bayesian or classical sensitivity analyses of over-parametrized models for incomplete categorical data, it is well known that prior-dependence on posterior inferences of nonidentifiable parameters or that too parsimonious over-parametrized models may lead to erroneous conclusions. Nevertheless, some authors either pay no attention to which parameters are nonidentifiable or do not appropriately account for possible prior-dependence. We review the literature on this topic and consider simple examples to emphasize that in both inferential frameworks, the subjective components can influence results in nontrivial ways, irrespectively of the sample size. Specifically, we show that prior distributions commonly regarded as slightly informative or noninformative may actually be too informative for nonidentifiable parameters, and that the choice of over-parametrized models may drastically impact the results, suggesting that a careful examination of their effects should be considered before drawing conclusions.Resume Que ce soit dans un cadre Bayesien ou classique, il est bien connu que la surparametrisation, dans les modeles pour donnees categorielles incompletes, peut conduire a des conclusions erronees. Cependant, certains auteurs persistent a negliger les problemes lies a la presence de parametres non identifies. Nous passons en revue la litterature dans ce domaine, et considerons quelques exemples surparametres simples dans lesquels les elements subjectifs influencent de facon non negligeable les resultats, independamment de la taille des echantillons. Plus precisement, nous montrons comment des a priori consideres comme peu ou non-informatifs peuvent se reveler extremement informatifs en ce qui concerne les parametres non identifies, et que le recours a des modeles surparametres peut avoir sur les conclusions finales un impact considerable. Ceci suggere un examen tres attentif de l`impact potentiel des a priori.

Real-Time Recognition System for Traffic Signs

The aim of this thesis project is to develop the Traffic Sign Recognition algorithm for real time. Inreal time environment, vehicles move at high speed on roads. For the vehicle intelligent system itbecomes essential to detect, process and recognize the traffic sign which is coming in front ofvehicle with high relative velocity, at the right time, so that the driver would be able to pro-actsimultaneously on instructions given in the Traffic Sign. The system assists drivers about trafficsigns they did not recognize before passing them. With the Traffic Sign Recognition system, thevehicle becomes aware of the traffic environment and reacts according to the situation.The objective of the project is to develop a system which can recognize the traffic signs in real time.The three target parameters are the system’s response time in real-time video streaming, the trafficsign recognition speed in still images and the recognition accuracy. The system consists of threeprocesses; the traffic sign detection, the traffic sign recognition and the traffic sign tracking. Thedetection process uses physical properties of traffic signs based on a priori knowledge to detect roadsigns. It generates the road sign image as the input to the recognition process. The recognitionprocess is implemented using the Pattern Matching algorithm. The system was first tested onstationary images where it showed on average 97% accuracy with the average processing time of0.15 seconds for traffic sign recognition. This procedure was then applied to the real time videostreaming. Finally the tracking of traffic signs was developed using Blob tracking which showed theaverage recognition accuracy to 95% in real time and improved the system’s average response timeto 0.04 seconds. This project has been implemented in C-language using the Open Computer VisionLibrary.

Financial time series analysis : Chaos and neurodynamics approach

This work aims at combining the Chaos theory postulates and Artificial Neural Networks classification and predictive capability, in the field of financial time series prediction. Chaos theory, provides valuable qualitative and quantitative tools to decide on the predictability of a chaotic system. Quantitative measurements based on Chaos theory, are used, to decide a-priori whether a time series, or a portion of a time series is predictable, while Chaos theory based qualitative tools are used to provide further observations and analysis on the predictability, in cases where measurements provide negative answers. Phase space reconstruction is achieved by time delay embedding resulting in multiple embedded vectors. The cognitive approach suggested, is inspired by the capability of some chartists to predict the direction of an index by looking at the price time series. Thus, in this work, the calculation of the embedding dimension and the separation, in Takens‘ embedding theorem for phase space reconstruction, is not limited to False Nearest Neighbor, Differential Entropy or other specific method, rather, this work is interested in all embedding dimensions and separations that are regarded as different ways of looking at a time series by different chartists, based on their expectations. Prior to the prediction, the embedded vectors of the phase space are classified with Fuzzy-ART, then, for each class a back propagation Neural Network is trained to predict the last element of each vector, whereas all previous elements of a vector are used as features.

Le long trajet vers le Nord : de la traduction de la littérature de jeunesse française en suédois

Depuis très récemment, un changement est en train de se produire dans le domaine de la traduction de la littérature jeunesse: auparavant et pendant de nombreuses années, les ouvrages traduits venaient le plus souvent de la Suède pour être traduits en français (Gossas & Lindgren, 2011). Dans l’autre sens, on trouvait des imagiers plus ou moins sans texte, des bandes-dessinées (surtout franco-belges), des retraductions et quelques classiques comme les très populaires ouvrages de Jules Verne ou Le Petit Prince. Or, la littérature jeunesse française est certes abondante – et ce même si l’on s’en tient aux livres publiés en France, mais aussi souvent de grande qualité littéraire et iconique, ce qui devrait a priori encourager sa diffusion. Cette littérature a de surcroît la chance de pouvoir surfer sur la vague actuelle des traductions du français au suédois de livres ne faisant pas partie de cette catégorie. Mais il est bien connu que le système littéraire suédois de la littérature jeunesse comporte peu de livres traduits, en dehors des traductions massives d’ouvrages en anglais. Les questions auxquelles nous répondrons lors de notre présentation sont : Quels sont les livres actuellement traduits du français vers le suédois ? Qui les traduit (quels éditeurs et quels traducteurs) ? Nous allons ainsi montrer quelle a été l’évolution depuis 1970 jusqu’à nos jours. Nous ferons aussi le lien avec nos études précédentes concernant le sens inverse, c’est-à-dire la traduction en français de la littérature suédoise contemporaine pour la jeunesse et de nos résultats en la matière sur les livres traduits, les éditeurs et les traducteurs. Nous utiliserons comme base théorique la théorie du polysystème (Even-Zohar), notamment concernant le rôle de la littérature traduite dans un système littéraire, et les études descriptives de la traduction (voir Toury, Shavit, etc.).

Measuring refugee malnutrition in the under-fives: pitfalls in practice

This article reviews the merit of using the World Health Organization (WHO) / National Center for Health Statistics (NCHS) references to evaluate the nutritional status of refugees and famine-affected populations and the implication of the findings for further development of Public Health Nutrition programs. Public health nutrition interventions, in order to change the nutritional status of refugees and famine affected-populations, need to be based on accurate and appropriate reference criteria in order to satisfy demands for cost-effectiveness and preserve the human right to adequate food. However, many are confused by the current practice of estimating malnutrition in complex emergencies and it appears that the use of WHO/NCHS references may not be appropriate for use under these conditions in particular, and in the third world in general. To verify this claim, we conducted a literature review. Factors to be reviewed were selected a priori. Computer-assisted searches for English-language publications in the MEDLINE database were conducted and references cited in related publications were reviewed. The literature was restricted to published papers from 1975 to 2000, which focused on malnutrition and its measurement. Findings from the literature review indicate that there is a need to develop more appropriate references that are reflective of the field of intervention. In the absence of more appropriate references, this paper presents the advantages and disadvantages of using these references and some policy recommendations.

What are we looking at, and how big is it?

Some of the most important outcomes of physical therapy treatment have to do with behaviour and quality of life. This article involves examining what it is we are measuring in physical therapy research and what those measurements mean. In looking at differences between groups (e.g. placebo-control) or strength of association between variables (e.g. correlation, regression) the practitioner/researcher must consider what are meaningful magnitudes of effects. Depending on the variable that one measures, a medium effect size (e.g. Cohen's d=0.50) may, in the real world, be insignificant, or in the case of elite athletic performance such an effect size might be gigantic. A major problem in the sports sciences is the confusion of p values and significance testing with the results of interest, the magnitudes of effects. Also, the prevalence of possible Type II errors in the sports sciences and medicine may be quite high in light of the small sample sizes and the paucity of power analyses for non-significant results. We make an appeal for determining a priori minimal meaningful differences (or associations) to use as the primary metrics in discussing results.

Multiresolution analysis of connectivity

Multiresolution histograms have been used for indexing and retrieval of images. Multiresolution histograms used traditionally are 2d-histograms which encode pixel intensities. Earlier we proposed a method for decomposing images by connectivity. In this paper, we propose to encode centroidal distances of an image in multiresolution histograms; the image is decomposed a priori, by connectivity. Multiresolution histograms thus obtained are 3d-histograms which encode connectivity and centroidal distances. The statistical technique of Principal Component Analysis is applied to multiresolution 3d-histograms and the resulting data is used to index images. Distance between two images is computed as the L2-difference of their principal components. Experiments are performed on Item S8 within the MPEG-7 image dataset. We also analyse the effect of pixel intensity thresholding on multiresolution images.

A trial of nurse practitioner scope of practice

Aims. The aim of this paper is to report a trial to investigate the feasibility of the nurse practitioner role in local health service delivery and to provide information about the educational and legislative requirements for nurse practitioner practice.

Background. Nurse practitioners have been shown to offer a beneficial service and fill a gap in health care provision. However, the lack of publications describing, critiquing, or defending the way that existing nurse practitioner roles have been developed may lead to a lack of clarity in comparing the nurse practitioner scope of practice internationally. In Australia, credible exploratory research is needed to realize the potential of nurse practitioners to bridge the divide of inequitable distribution of health services. A trial of nurse practitioner services in the Australian Capital Territory provided an excellent opportunity to investigate these scope and continuity issues.

Methods. This was an observational analytic study using multiple data sources. Four models of nurse practitioner service were chosen from a competitive field of applications that were evaluated according to efficacy, feasibility, and sustainability across specified selection criteria. Each model in the trial included a clinical support team, with the nurse practitioner candidate 'working-into-the-role' and collecting demographic, clinical practice, patient outcome, and health service and consumer survey data over a 10 month period.

Findings. The trial identified the broad potential of the nurse practitioner role, its breadth and limitations, and its impact on selected health services in the Australian Capital Territory. Data from individual models were compared highlighting generic elements, and formed the basis for the development of the scope of practice for the Australian Capital Territory nurse practitioner models.

Conclusions. This study has validated a research-based, iterative process for initial development of nurse practitioner scope of practice for any Australian specialization. Importantly, the study concluded with the scope of practice as a finding, rather than commencing with it a priori. Although general areas of health care need and under-servicing were identified at the outset, the process tested both the expansion and parameters of the roles.

Why does the universe exist? An advaita vedantic perspective

Debates concerning causal explanations of the universe tend to be based on a priori propositions (e.g., Edwards, 1973; Smith, 1995; Swinburne, 1978). The present paper, however, addresses the metaphysical question, “Why does the universe exist?” from the perspective of a school of Hindu philosophy referred to as advaita vedanta and two of its a posteriori derived creation theories: the theory of simultaneous creation (drishti-srishti vada) and the theory of non-causality (ajata vada). Objections to advaita vedanta are also discussed. It is concluded that advaita vedanta has the potential to make a significant contribution to contemporary metaphysical debate in general and our understanding of the question, “Why does the universe exist?” in particular.

Interventions for preventing obesity in childhood. A systematic review

Background The prevalence of obesity and overweight is increasing worldwide. Obesity in children impacts on their health in both short- and long-term. Obesity prevention strategies are poorly understood.

Objective To assess the effectiveness of interventions designed to prevent obesity in childhood.

Search strategy Electronic databases were searched from January 1985 to October 1999.

Selection criteria Data from randomized control trials and non-randomized trials with concurrent control group were included. A priori, studies with follow up of 1 year minimum were selected however, this was subsequently amended to include studies with a minimum follow up of three months.

Data collection & analysis Two reviewers independently extracted data and assessed study quality.

Main results Seven studies were included, three long-term (>1 years) and four short-term (>3 months and <1 years). The studies included were diverse in terms of study design and quality, target population, theoretical underpinning of intervention approach, and outcome measures. As such, it was not appropriate to combine study findings using statistical methods.

Conclusions Two of the long-term studies (one focused on dietary education and physical activity vs. control, and the other only on dietary education vs. control), resulted in a reduction in the prevalence on obesity, but the third, which focused on dietary education and physical activity, found no effect. Of the four short-term studies, three focused simply on physical activity/reduction of sedentary behavious vs. control. Two of these studies resulted in a reduction in the prevalence of obesity in intervention groups compared with control groups, and another study found a non-significant reduction. The fourth study focused on dietary education and physical activity, and did not find an effect on obesity, but did report a reduction in fat intake. Overall, the findings of the review suggest that currently there is limited quality data on the effectiveness of obesity prevention programmes and as such no generalizable conclusions can be drawn. The need for well-designed studies that examine a range of interventions remains a priority.

Should older people in residential care receive vitamin D to prevent falls? Results of a randomized trial

Objectives: To determine whether vitamin D supplementation can reduce the incidence of falls and fractures in older people in residential care who are not classically vitamin D deficient.

Design: Randomized, placebo-controlled double-blind, trial of 2 years' duration.

Setting: Multicenter study in 60 hostels (assisted living facilities) and 89 nursing homes across Australia.

Participants: Six hundred twenty-five residents (mean age 83.4) with serum 25-hydroxyvitamin D levels between 25 and 90 nmol/L.

Intervention: Vitamin D supplementation (ergocalciferol, initially 10,000 IU given once weekly and then 1,000 IU daily) or placebo for 2 years. All subjects received 600 mg of elemental calcium daily as calcium carbonate.

Measurements: Falls and fractures recorded prospectively in study diaries by care staff.

Results: The vitamin D and placebo groups had similar baseline characteristics. In intention-to-treat analysis, the incident rate ratio for falling was 0.73 (95% confidence interval (CI)=0.57&ndash;0.95). The odds ratio for ever falling was 0.82 (95% CI=0.59&ndash;1.12) and for ever fracturing was 0.69 (95% CI=0.40&ndash;1.18). An a priori subgroup analysis of subjects who took at least half the prescribed capsules (n=540), demonstrated an incident rate ratio for falls of 0.63 (95% CI=0.48&ndash;0.82), an odds ratio (OR) for ever falling of 0.70 (95% CI=0.50&ndash;0.99), and an OR for ever fracturing of 0.68 (95% CI=0.38&ndash;1.22).

Conclusion: Older people in residential care can reduce their incidence of falls if they take a vitamin D supplement for 2 years even if they are not initially classically vitamin D deficient.

The relative importance of landscape properties for woodland birds in agricultural environments

1. Studies of landscape change are seldom conducted at scales commensurate with the processes they purport to investigate. Landscape change is a landscape-level process, yet most studies focus on patches. Even when landscape context is considered, inference remains at the patch-level. The unit of investigation must be extended beyond individual patches to whole mosaics in order to advance understanding of faunal responses to landscape change.

2. In this study, we aggregated data from multiple sites per landscape such that both the response and explanatory variables characterized 'whole' landscapes, allowing for landscape-level inference about factors influencing species' incidence.

3. We used hierarchical partitioning and Bayesian variable selection methods to develop species-specific models that examined the influence of four categories of landscape properties &ndash; habitat extent, habitat configuration, landscape composition and geographical location &ndash; on the incidence of 58 species of woodland-dependent birds in 24 agricultural landscapes (each 100 km2) in south-eastern Australia.

4. There was strong evidence for a positive effect of habitat extent for 27 species. Thirty species were related to at least one of the four landscape composition variables, and geographical location was important for 19 species. Habitat configuration was influential for 13 species and where important, the impacts of fragmentation per se were detrimental.

5. Variation among species in the influential landscape variables indicates that different species respond to different sets of cues in land mosaics. Thus, although all species were grouped a priori as 'woodland-dependent', expectations based on general ecological characteristics may prove unreliable.

6. Synthesis and applications. These results underscore the value of moving beyond the fragmentation paradigm focused on the spatial pattern of habitat vs. non-habitat, to a greater appreciation of the composition and heterogeneity of land mosaics. Landscape-level inference will enable improved conservation outcomes by recognizing the influence of landscape properties on biota and devising strategies at this scale to complement patch-based management. We provide strong empirical evidence that biodiversity management in agricultural landscapes must focus on habitat extent. Complementary management of other landscape attributes, such as habitat aggregation and intensity of agricultural land-use, will also enhance the value of agricultural landscapes for woodland birds.

Telling it how it really is?

Heideggerian phenomenology is expounded here as a qualitative methodology of value in researching "how things really are" in organizational life. It is a methodology, and a philosophy, that recognizes a priori the value of the subjective, lived experience of the individual, and the physical, sociological, and psychological phenomena that come together to create a life-world. In this paper, the authors present three cases where this methodology successfully revealed the employee's reality. Stories about working in teams, being downsized, and having an unseen chronic illness at work are shared to demonstrate the ability of Heideggerian phenomenology to reveal how life really is in organizations. The benefit for HR practitioners and managers is an increased understanding of and, hence, ability to address these sensitive and often unspoken aspects of organizational life.

Telling it how it really is? : life in organisations revealed using Heideggerian phenomenology

Heideggerian phenomenology is expounded here as a qualitative methodology of value in researching "how things really are" in organisational life. It is a methodology, and a philosophy, which recognises a priori the value ofthe subjective, lived experience of the informant, and the physical, sociological and psychological phenomena that come together to create their life-world. In this paper, the authors present cases where this methodology successfully revealed the respondents' reality. Stories about working in teams and being downsized are shared to demonstrate the ability of Heideggerian phenomenology to reveal how life really is in organisations.

Future research on structural and institutional forms of HIV discrimination

There is scant research into HIV discrimination, particularly structural and institutional forms of it. This appears, in part, to arise from a prevailing view that the solutions to HIV discrimination are known a priori and in the absence of additional data. The six country studies into HIV discrimination presented in this AIDS Care supplement suggest that this is not the case. There needs to be an acceptance that if the issue of HIV discrimination is to be tackled successfully and based on evidence, then incremental and effortful steps need to be taken to develop that evidence base. This is discussed.