Novel sequential methods in dose-finding designs : extensions of the continual reassessment method

Dose-finding trials are a form of clinical data collection process in which the primary objective is to estimate an optimum dose of an investigational new drug when given to a patient. This thesis develops and explores three novel dose-finding design methodologies. All design methodologies presented in this thesis are pragmatic. They use statistical models, incorporate clinicians' prior knowledge efficiently, and prematurely stop a trial for safety or futility reasons. Designing actual dose-finding trials using these methodologies will minimize practical difficulties, improve efficiency of dose estimation, be flexible to stop early and reduce possible patient discomfort or harm.

Promoting Statistical Literacy Through Data Modelling in the Early School Years

This chapter addresses data modelling as a means of promoting statistical literacy in the early grades. Consideration is first given to the importance of increasing young children’s exposure to statistical reasoning experiences and how data modelling can be a rich means of doing so. Selected components of data modelling are then reviewed, followed by a report on some findings from the third-year of a three-year longitudinal study across grades one through three.

Does counting still count? Revisiting the security of counting based user authentication protocols against statistical attacks

At NDSS 2012, Yan et al. analyzed the security of several challenge-response type user authentication protocols against passive observers, and proposed a generic counting based statistical attack to recover the secret of some counting based protocols given a number of observed authentication sessions. Roughly speaking, the attack is based on the fact that secret (pass) objects appear in challenges with a different probability from non-secret (decoy) objects when the responses are taken into account. Although they mentioned that a protocol susceptible to this attack should minimize this difference, they did not give details as to how this can be achieved barring a few suggestions. In this paper, we attempt to fill this gap by generalizing the attack with a much more comprehensive theoretical analysis. Our treatment is more quantitative which enables us to describe a method to theoretically estimate a lower bound on the number of sessions a protocol can be safely used against the attack. Our results include 1) two proposed fixes to make counting protocols practically safe against the attack at the cost of usability, 2) the observation that the attack can be used on non-counting based protocols too as long as challenge generation is contrived, 3) and two main design principles for user authentication protocols which can be considered as extensions of the principles from Yan et al. This detailed theoretical treatment can be used as a guideline during the design of counting based protocols to determine their susceptibility to this attack. The Foxtail protocol, one of the protocols analyzed by Yan et al., is used as a representative to illustrate our theoretical and experimental results.

Critical reflections on the use of participatory methodologies to build evaluation capacities in international development organisations

There has been a greater focus on strengthening evaluation capacity building (ECB) within development organisations in recent years. This can be attributed in part to the growing appreciation of the value of participatory and collaborative forms of evaluation. Evaluation is increasingly seen as an ongoing learning process and an important means of strengthening capacity and improving organisational performance (Horton et al., 2003:7). While there are many benefits of using participatory methodologies in ECB projects, our experiences and a review of the literature in this area highlight the many challenges, issues and contradictions that can affect the success of such ECB efforts. We discuss these issues, drawing on our learnings from the ongoing participatory action research (PAR) project 'Assessing Communication for Social Change’ (AC4SC). This four year project, which began in 2007, is a collaboration between communication and development academics and evaluation specialists from two Australian universities and communication for development practitioners and monitoring and evaluation (M&E) staff in the NGO Equal Access Nepal (EAN). The aim is to develop, implement, and evaluate a participatory methodology for assessing the social change impacts of community radio programs produced by EAN. It builds on previous projects that used ethnographic action research (EAR) methodology (Tacchi et al., 2007).

Critical reflections on the use of participatory methodologies to build evaluation capacities in international development organisations

"In this chapter the authors present a critique of Participatory Evaluation as worked in development projects, in this case, in Nepal. The article works between established claims that Participatory Evaluation builds capacity at programmatic and organisational levels, and the specific experiences of these claims in the authors’ current work. They highlight the need to address key difficulties such as high turn-over of staff and resulting loss of capacity to engage in Participatory Evaluation, and the difficulty of communication between academic as compared with local practical wisdoms. A key issue is the challenge of addressing the inevitable issues of power inequities that such approaches encounter. While Participatory Evaluation has been around for some time, it has only enjoyed more widespread recognition of its value in comparatively recent times, with its uptake in international development environments. To this extent, the practice is still in its early stages of development, and Jo, June and Michael’s work contributes to strengthening and more comprehensively understanding it. With regard to the meta-theme of this publication, this chapter is an example of how context not only influences the methodology to be used and the praxis of how it is to be used, but contributes to early explication of the core nature of an emerging methodology."

Statistical analysis of reduction in tensile strength of cotton strips as a measure of soil microbial activity

The cotton strip assay (CSA) is an established technique for measuring soil microbial activity. The technique involves burying cotton strips and measuring their tensile strength after a certain time. This gives a measure of the rotting rate, R, of the cotton strips. R is then a measure of soil microbial activity. This paper examines properties of the technique and indicates how the assay can be optimised. Humidity conditioning of the cotton strips before measuring their tensile strength reduced the within and between day variance and enabled the distribution of the tensile strength measurements to approximate normality. The test data came from a three-way factorial experiment (two soils, two temperatures, three moisture levels). The cotton strips were buried in the soil for intervals of time ranging up to 6 weeks. This enabled the rate of loss of cotton tensile strength with time to be studied under a range of conditions. An inverse cubic model accounted for greater than 90% of the total variation within each treatment combination. This offers support for summarising the decomposition process by a single parameter R. The approximate variance of the decomposition rate was estimated from a function incorporating the variance of tensile strength and the differential of the function for the rate of decomposition, R, with respect to tensile strength. This variance function has a minimum when the measured strength is approximately 2/3 that of the original strength. The estimates of R are almost unbiased and relatively robust against the cotton strips being left in the soil for more or less than the optimal time. We conclude that the rotting rate X should be measured using the inverse cubic equation, and that the cotton strips should be left in the soil until their strength has been reduced to about 2/3.

Design methodologies and architectures of hardware-based evolutionary algorithms for aerospace optimisation applications on FPGAS

This thesis is a study of new design methods for allowing evolutionary algorithms to be more effectively utilised in aerospace optimisation applications where computation needs are high and computation platform space may be restrictive. It examines the applicability of special hardware computational platforms known as field programmable gate arrays and shows that with the right implementation methods they can offer significant benefits. This research is a step forward towards the advancement of efficient and highly automated aircraft systems for meeting compact physical constraints in aerospace platforms and providing effective performance speedups over traditional methods.

Field validation of controlled Monte Carlo data generation for statistical damage identification employing Mahalanobis squared distance

This article presents the field applications and validations for the controlled Monte Carlo data generation scheme. This scheme was previously derived to assist the Mahalanobis squared distance–based damage identification method to cope with data-shortage problems which often cause inadequate data multinormality and unreliable identification outcome. To do so, real-vibration datasets from two actual civil engineering structures with such data (and identification) problems are selected as the test objects which are then shown to be in need of enhancement to consolidate their conditions. By utilizing the robust probability measures of the data condition indices in controlled Monte Carlo data generation and statistical sensitivity analysis of the Mahalanobis squared distance computational system, well-conditioned synthetic data generated by an optimal controlled Monte Carlo data generation configurations can be unbiasedly evaluated against those generated by other set-ups and against the original data. The analysis results reconfirm that controlled Monte Carlo data generation is able to overcome the shortage of observations, improve the data multinormality and enhance the reliability of the Mahalanobis squared distance–based damage identification method particularly with respect to false-positive errors. The results also highlight the dynamic structure of controlled Monte Carlo data generation that makes this scheme well adaptive to any type of input data with any (original) distributional condition.

Multivariate statistical analysis of hydrochemical data to assess alluvial aquifer–stream connectivity during drought and flood : Cressbrook Creek, southeast Queensland, Australia

A catchment-scale multivariate statistical analysis of hydrochemistry enabled assessment of interactions between alluvial groundwater and Cressbrook Creek, an intermittent drainage system in southeast Queensland, Australia. Hierarchical cluster analyses and principal component analysis were applied to time-series data to evaluate the hydrochemical evolution of groundwater during periods of extreme drought and severe flooding. A simple three-dimensional geological model was developed to conceptualise the catchment morphology and the stratigraphic framework of the alluvium. The alluvium forms a two-layer system with a basal coarse-grained layer overlain by a clay-rich low-permeability unit. In the upper and middle catchment, alluvial groundwater is chemically similar to streamwater, particularly near the creek (reflected by high HCO3/Cl and K/Na ratios and low salinities), indicating a high degree of connectivity. In the lower catchment, groundwater is more saline with lower HCO3/Cl and K/Na ratios, notably during dry periods. Groundwater salinity substantially decreased following severe flooding in 2011, notably in the lower catchment, confirming that flooding is an important mechanism for both recharge and maintaining groundwater quality. The integrated approach used in this study enabled effective interpretation of hydrological processes and can be applied to a variety of hydrological settings to synthesise and evaluate large hydrochemical datasets.

A statistical approach to quantify the impact on voltage quality caused by PV generators

A probabilistic method is proposed to evaluate voltage quality of grid-connected photovoltaic (PV) power systems. The random behavior of solar irradiation is described in statistical terms and the resulting voltage fluctuation probability distribution is then derived. Reactive power capabilities of the PV generators are then analyzed and their operation under constant power factor mode is examined. By utilizing the reactive power capability of the PV-generators to the full, it is shown that network voltage quality can be greatly enhanced.