Classifier selection using sequential error ratio criterion for multi-instance and multi-sample fusion

Classifier selection is a problem encountered by multi-biometric systems that aim to improve performance through fusion of decisions. A particular decision fusion architecture that combines multiple instances (n classifiers) and multiple samples (m attempts at each classifier) has been proposed in previous work to achieve controlled trade-off between false alarms and false rejects. Although analysis on text-dependent speaker verification has demonstrated better performance for fusion of decisions with favourable dependence compared to statistically independent decisions, the performance is not always optimal. Given a pool of instances, best performance with this architecture is obtained for certain combination of instances. Heuristic rules and diversity measures have been commonly used for classifier selection but it is shown that optimal performance is achieved for the `best combination performance' rule. As the search complexity for this rule increases exponentially with the addition of classifiers, a measure - the sequential error ratio (SER) - is proposed in this work that is specifically adapted to the characteristics of sequential fusion architecture. The proposed measure can be used to select a classifier that is most likely to produce a correct decision at each stage. Error rates for fusion of text-dependent HMM based speaker models using SER are compared with other classifier selection methodologies. SER is shown to achieve near optimal performance for sequential fusion of multiple instances with or without the use of multiple samples. The methodology applies to multiple speech utterances for telephone or internet based access control and to other systems such as multiple finger print and multiple handwriting sample based identity verification systems.

Design characteristic of randomised controlled trials for geographic atrophy in age-related macular degeneration: selection of outcomes and sample size calculation

PurposeThe selection of suitable outcomes and sample size calculation are critical factors in the design of a randomised controlled trial (RCT). The goal of this study was to identify the range of outcomes and information on sample size calculation in RCTs on geographic atrophy (GA).MethodsWe carried out a systematic review of age-related macular degeneration (AMD) RCTs. We searched MEDLINE, EMBASE, Scopus, Cochrane Library, www.controlled-trials.com, and www.ClinicalTrials.gov. Two independent reviewers screened records. One reviewer collected data and the second reviewer appraised 10% of collected data. We scanned references lists of selected papers to include other relevant RCTs.ResultsLiterature and registry search identified 3816 abstracts of journal articles and 493 records from trial registries. From a total of 177 RCTs on all types of AMD, 23 RCTs on GA were included. Eighty-one clinical outcomes were identified. Visual acuity (VA) was the most frequently used outcome, presented in 18 out of 23 RCTs and followed by the measures of lesion area. For sample size analysis, 8 GA RCTs were included. None of them provided sufficient Information on sample size calculations.ConclusionsThis systematic review illustrates a lack of standardisation in terms of outcome reporting in GA trials and issues regarding sample size calculation. These limitations significantly hamper attempts to compare outcomes across studies and also perform meta-analyses.

Finite-Sample Simulation-Based Inference in VAR Models with Applications to Order Selection and Causality Testing

Statistical tests in vector autoregressive (VAR) models are typically based on large-sample approximations, involving the use of asymptotic distributions or bootstrap techniques. After documenting that such methods can be very misleading even with fairly large samples, especially when the number of lags or the number of equations is not small, we propose a general simulation-based technique that allows one to control completely the level of tests in parametric VAR models. In particular, we show that maximized Monte Carlo tests [Dufour (2002)] can provide provably exact tests for such models, whether they are stationary or integrated. Applications to order selection and causality testing are considered as special cases. The technique developed is applied to quarterly and monthly VAR models of the U.S. economy, comprising income, money, interest rates and prices, over the period 1965-1996.

Validity of an experimental underwater demolition team selection battery for an officer sample /

Bibliographical footnotes.

Optimal Budget-Constrained Sample Allocation for Selection Decisions with Multiple Uncertain Attributes

A decision-maker, when faced with a limited and fixed budget to collect data in support of a multiple attribute selection decision, must decide how many samples to observe from each alternative and attribute. This allocation decision is of particular importance when the information gained leads to uncertain estimates of the attribute values as with sample data collected from observations such as measurements, experimental evaluations, or simulation runs. For example, when the U.S. Department of Homeland Security must decide upon a radiation detection system to acquire, a number of performance attributes are of interest and must be measured in order to characterize each of the considered systems. We identified and evaluated several approaches to incorporate the uncertainty in the attribute value estimates into a normative model for a multiple attribute selection decision. Assuming an additive multiple attribute value model, we demonstrated the idea of propagating the attribute value uncertainty and describing the decision values for each alternative as probability distributions. These distributions were used to select an alternative. With the goal of maximizing the probability of correct selection we developed and evaluated, under several different sets of assumptions, procedures to allocate the fixed experimental budget across the multiple attributes and alternatives. Through a series of simulation studies, we compared the performance of these allocation procedures to the simple, but common, allocation procedure that distributed the sample budget equally across the alternatives and attributes. We found the allocation procedures that were developed based on the inclusion of decision-maker knowledge, such as knowledge of the decision model, outperformed those that neglected such information. Beginning with general knowledge of the attribute values provided by Bayesian prior distributions, and updating this knowledge with each observed sample, the sequential allocation procedure performed particularly well. These observations demonstrate that managing projects focused on a selection decision so that the decision modeling and the experimental planning are done jointly, rather than in isolation, can improve the overall selection results.

Attention and memory bias for body image and health related information using an Emotional Stroop task in a non-clinical sample

It has been proposed that body image disturbance is a form of cognitive bias wherein schemas for self-relevant information guide the selective processing of appearancerelated information in the environment. This threatening information receives disproportionately more attention and memory, as measured by an Emotional Stroop and incidental recall task. The aim of this thesis was to expand the literature on cognitive processing biases in non-clinical males and females by incorporating a number of significant methodological refinements. To achieve this aim, three phases of research were conducted. The initial two phases of research provided preliminary data to inform the development of the main study. Phase One was a qualitative exploration of body image concerns amongst males and females recruited through the general community and from a university. Seventeen participants (eight male; nine female) provided information on their body image and what factors they saw as positively and negatively impacting on their self evaluations. The importance of self esteem, mood, health and fitness, and recognition of the social ideal were identified as key themes. These themes were incorporated as psycho-social measures and Stroop word stimuli in subsequent phases of the research. Phase Two involved the selection and testing of stimuli to be used in the Emotional Stroop task. Six experimental categories of words were developed that reflected a broad range of health and body image concerns for males and females. These categories were high and low calorie food words, positive and negative appearance words, negative emotion words, and physical activity words. Phase Three addressed the central aim of the project by examining cognitive biases for body image information in empirically defined sub-groups. A National sample of males (N = 55) and females (N = 144), recruited from the general community and universities, completed an Emotional Stroop task, incidental memory test, and a collection of psycho-social questionnaires. Sub-groups of body image disturbance were sought using a cluster analysis, which identified three sub-groups in males (Normal, Dissatisfied, and Athletic) and four sub-groups in females (Normal, Health Conscious, Dissatisfied, and Symptomatic). No differences were noted between the groups in selective attention, although time taken to colour name the words was associated with some of the psycho-social variables. Memory biases found across the whole sample for negative emotion, low calorie food, and negative appearance words were interpreted as reflecting the current focus on health and stigma against being unattractive. Collectively these results have expanded our understanding of processing biases in the general community by demonstrating that the processing biases are found within non-clinical samples and that not all processing biases are associated with negative functionality

An investigation of building physical properties selection to combat global warming

Global warming can have a significant impact on building energy performance and indoor thermal environment, as well as the health and productivity of people living and working inside them. Through the building simulation technique, this paper investigates the adaptation potential of different selections of building physical properties to increased outdoor temperature in Australia. It is found that overall, an office building with lower insulation level, smaller window to wall ratio and/or a glass type with lower shading coefficient, and lower internal load density will have the effect of lowering building cooling load and total energy use, and therefore have a better potential to adapt to the warming external climate. Compared with clear glass, it is shown that the use of reflective glass for the sample building with WWR being 0.5 reduces the building cooling load by more than 12%. A lower internal load can also have a significant impact on the reduction of building cooling load, as well as the building energy use. Through the comparison of results between current and future weather scenarios, it is found that the patterns found in the current weather scenario also exist in the future weather scenarios, but to a smaller extent.

Wheat breeding nurseries, target environments, and indirect selection for grain yield

Use of appropriate nursery environments will maximize gain from selection for yield of wheat (Triticum aestivum L.) in the target population of environments of a breeding program. The objective of this study was to investigate how well-irrigated (low-stress) nursery environments predict yield of lines in target environments that varied in degree of water limitation. Fifteen lines were sampled from the preliminary yield evaluation stage of the Queensland wheat breeding program and tested in 26 trials under on-farm conditions (Target Environments) across nine years (1985 to 1993) and also in 27 trials conducted at three research stations (Nursery Environments) in three years (1987 to 1989). The nursery environments were structured to impose different levels of water and nitrogen (N) limitation, whereas the target environments represented a random sample of on-farm conditions from the target population of environments. Indirect selection and pattern analysis methods were used to investigate selection for yield in the nursery environments and gain from selection in the target environments. Yield under low-stress nursery conditions was an effective predictor of yield under similar low-stress target environments (r = 0.89, P < 0.01). However, the value of the low-stress nursery as a predictor of yield in the water-limited target environments decreased with increasing water stress (moderate stress r = 0.53, P < 0.05, to r = 0.38, P > 0.05; severe stress r = -0.08, P > 0.05). Yield in the stress nurseries was a poor predictor of yield in the target environments. Until there is a clear understanding of the physiological-genetic basis of variation for adaptation of wheat to the water-limited environments in Queensland, yield improvement can best be achieved by selection for a combination of yield potential in an irrigated low-stress nursery and yield in on-farm trials that sample the range of water-limited environments of the target population of environments.

The feeding practices and structure questionnaire : construction and initial validation in a sample of Australian first-time mothers and their 2-year olds

Background Early feeding practices lay the foundation for children’s eating habits and weight gain. Questionnaires are available to assess parental feeding but overlapping and inconsistent items, subscales and terminology limit conceptual clarity and between study comparisons. Our aim was to consolidate a range of existing items into a parsimonious and conceptually robust questionnaire for assessing feeding practices with very young children (<3 years). Methods Data were from 462 mothers and children (age 21–27 months) from the NOURISH trial. Items from five questionnaires and two study-specific items were submitted to a priori item selection, allocation and verification, before theoretically-derived factors were tested using Confirmatory Factor Analysis. Construct validity of the new factors was examined by correlating these with child eating behaviours and weight. Results Following expert review 10 factors were specified. Of these, 9 factors (40 items) showed acceptable model fit and internal reliability (Cronbach’s α: 0.61-0.89). Four factors reflected non-responsive feeding practices: ‘Distrust in Appetite’, ‘Reward for Behaviour’, ‘Reward for Eating’, and ‘Persuasive Feeding’. Five factors reflected structure of the meal environment and limits: ‘Structured Meal Setting’, ‘Structured Meal Timing’, ‘Family Meal Setting’, ‘Overt Restriction’ and ‘Covert Restriction’. Feeding practices generally showed the expected pattern of associations with child eating behaviours but none with weight. Conclusion The Feeding Practices and Structure Questionnaire (FPSQ) provides a new reliable and valid measure of parental feeding practices, specifically maternal responsiveness to children’s hunger/satiety signals facilitated by routine and structure in feeding. Further validation in more diverse samples is required.

Who selects the 'right' directors? An examination of the association between board selection, gender diversity and outcomes

Using a sample of companies from the top 500 listed firms in Australia, we investigate whether the presence of a designated nomination committee and female representation on the nomination committee affect board gender diversity. We also examine whether gender diversity on the board affects firm risk and financial performance. We find that board gender diversity is significantly and positively associated with the presence of a designated nomination committee and that female representation on the nomination committee is a significant explanatory factor of increasing board gender diversity following the release of the 2010 Australian Securities Exchange Corporate Governance Council (ASXCGC) recommendations. Further, our results support the business case for board gender diversity as we find greater gender diversity moderates excessive firm risk which in turn improves firms’ financial performance. Our results are robust after correcting for selection bias and controlling for other board, firm and industry characteristics.

The on-road experiences and awareness of sleepiness in a sample of Australian highway drivers: A roadside driver sleepiness study

Objective Driver sleepiness contributes substantially to road crash incidents. Simulator and on-road studies clearly reveal an impairing effect from sleepiness on driving ability. However, the degree to which drivers appreciate the dangerousness of driving while sleepy is somewhat unclear. This study sought to determine drivers’ on-road experiences of sleepiness, their prior sleep habits, and personal awareness of the signs of sleepiness. Methods Participants were a random selection of 92 drivers travelling on a major highway in the state of Queensland, Australia, who were stopped by police as part of routine drink driving operations. Participants completed a brief questionnaire that included demographic information, sleepy driving experiences (signs of sleepiness and on-road experiences of sleepiness), and prior sleep habits. A modified version of the Karolinska Sleepiness Scale (KSS) was used to assess subjective sleepiness in the 15 minutes prior to being stopped by police. Results Participants rating of subjective sleepiness were quite low, with 90% reporting being alert to extremely alert on the KSS. Participants were reasonably aware of the signs of sleepiness, with many signs of sleepiness associated with on-road experiences of sleepiness. Additionally, the number of hours spent driving was positively correlated with the drivers’ level of sleep debt. Conclusions The results suggest the participants had moderate experience of driving while sleepy and many were aware of the signs of sleepiness. The relationship between driving long distances and increased sleep debt is a concern for road safety – increased education regarding the dangers of sleepy driving seems warranted.

Genetic analyses in a sample of individuals with high or low BMD shows association with multiple Wnt pathway genes

INTRODUCTION Although the high heritability of BMD variation has long been established, few genes have been conclusively shown to affect the variation of BMD in the general population. Extreme truncate selection has been proposed as a more powerful alternative to unselected cohort designs in quantitative trait association studies. We sought to test these theoretical predictions in studies of the bone densitometry measures BMD, BMC, and femoral neck area, by investigating their association with members of the Wnt pathway, some of which have previously been shown to be associated with BMD in much larger cohorts, in a moderate-sized extreme truncate selected cohort (absolute value BMD Z-scores = 1.5-4.0; n = 344). MATERIALS AND METHODS Ninety-six tag-single nucleotide polymorphism (SNPs) lying in 13 Wnt signaling pathway genes were selected to tag common genetic variation (minor allele frequency [MAF] > 5% with an r(2) > 0.8) within 5 kb of all exons of 13 Wnt signaling pathway genes. The genes studied included LRP1, LRP5, LRP6, Wnt3a, Wnt7b, Wnt10b, SFRP1, SFRP2, DKK1, DKK2, FZD7, WISP3, and SOST. Three hundred forty-four cases with either high or low BMD were genotyped by Illumina Goldengate microarray SNP genotyping methods. Association was tested either by Cochrane-Armitage test for dichotomous variables or by linear regression for quantitative traits. RESULTS Strong association was shown with LRP5, polymorphisms of which have previously been shown to influence total hip BMD (minimum p = 0.0006). In addition, polymorphisms of the Wnt antagonist, SFRP1, were significantly associated with BMD and BMC (minimum p = 0.00042). Previously reported associations of LRP1, LRP6, and SOST with BMD were confirmed. Two other Wnt pathway genes, Wnt3a and DKK2, also showed nominal association with BMD. CONCLUSIONS This study shows that polymorphisms of multiple members of the Wnt pathway are associated with BMD variation. Furthermore, this study shows in a practical trial that study designs involving extreme truncate selection and moderate sample sizes can robustly identify genes of relevant effect sizes involved in BMD variation in the general population. This has implications for the design of future genome-wide studies of quantitative bone phenotypes relevant to osteoporosis.

Model selection with misspecified spatial covariance structure

Spatial data analysis has become more and more important in the studies of ecology and economics during the last decade. One focus of spatial data analysis is how to select predictors, variance functions and correlation functions. However, in general, the true covariance function is unknown and the working covariance structure is often misspecified. In this paper, our target is to find a good strategy to identify the best model from the candidate set using model selection criteria. This paper is to evaluate the ability of some information criteria (corrected Akaike information criterion, Bayesian information criterion (BIC) and residual information criterion (RIC)) for choosing the optimal model when the working correlation function, the working variance function and the working mean function are correct or misspecified. Simulations are carried out for small to moderate sample sizes. Four candidate covariance functions (exponential, Gaussian, Matern and rational quadratic) are used in simulation studies. With the summary in simulation results, we find that the misspecified working correlation structure can still capture some spatial correlation information in model fitting. When the sample size is large enough, BIC and RIC perform well even if the the working covariance is misspecified. Moreover, the performance of these information criteria is related to the average level of model fitting which can be indicated by the average adjusted R square ( [GRAPHICS] ), and overall RIC performs well.

Modeling strategies in longitudinal data analysis: Covariate, variance function and correlation structure selection

A modeling paradigm is proposed for covariate, variance and working correlation structure selection for longitudinal data analysis. Appropriate selection of covariates is pertinent to correct variance modeling and selecting the appropriate covariates and variance function is vital to correlation structure selection. This leads to a stepwise model selection procedure that deploys a combination of different model selection criteria. Although these criteria find a common theoretical root based on approximating the Kullback-Leibler distance, they are designed to address different aspects of model selection and have different merits and limitations. For example, the extended quasi-likelihood information criterion (EQIC) with a covariance penalty performs well for covariate selection even when the working variance function is misspecified, but EQIC contains little information on correlation structures. The proposed model selection strategies are outlined and a Monte Carlo assessment of their finite sample properties is reported. Two longitudinal studies are used for illustration.

Selection of Corymbia citriodora for pulp productivity

Evaluation of a series of spotted gum (Corymbia citirodora) progeny trials, established in the subtropical region of Queensland, Australia, was undertaken to provide information for the development of advanced-generation breeding populations suitable for pulp production. Measurements of growth at two ages were combined with assessments of wood density and pulp yield from a selected sample of provenances to provide comparisons between provenances, to generate genetic parameter estimates and to predict genetic gain potential. Although growth at this age was moderate relative to other eucalypts, the near-infrared predictions of average wood density of 756 kg m(-3) and pulp yield of 55% indicate the species has considerable potential as a pulpwood crop. A pulp productivity breeding objective was used to identify production populations using a range of selection trait weightings to determine potential genetic gain for pulp productivity. Genetic parameters indicated (1) levels of genetic control were moderate for all traits and higher for wood property traits, (2) genetic improvements could be achieved by selection among and within provenances with greater levels of improvement available from selection within populations, (3) genotype by environment interactions were negligible, (4) genetic correlations between traits were favourable, and (5) selection of volume production alone would maximise improvements in pulp productivity.