arXiv subjects

Kerrie Mengersen

Publications and source records attributed to Kerrie Mengersen.

At least 55 records · Page 3Linked to original sources

arXiv2022

Bayesian Design with Sampling Windows for Complex Spatial Processes

Optimal design facilitates intelligent data collection. In this paper, we introduce a fully Bayesian design approach for spatial processes with complex covariance structures, like those typically exhibited in natural ecosystems. Coordinate Exchange algorithms are commonly used to find optimal design points. However, collecting data at specific points is often infeasible in practice. Currently, there is no provision to allow for flexibility in the choice of design. We also propose an approach to find Bayesian sampling windows, rather than points, via Gaussian process emulation to identify regions of high design efficiency across a multi-dimensional space. These developments are motivated by two ecological case studies: monitoring water temperature in a river network system in the northwestern United States and monitoring submerged coral reefs off the north-west coast of Australia.

stat.ME↗

arXiv2022

Stateful to Stateless: Modelling Stateless Ethereum

The concept of 'Stateless Ethereum' was conceived with the primary aim of mitigating Ethereum's unbounded state growth. The key facilitator of Stateless Ethereum is through the introduction of 'witnesses' into the ecosystem. The changes and potential consequences that these additional data packets pose on the network need to be identified and analysed to ensure that the Ethereum ecosystem can continue operating securely and efficiently. In this paper we propose a Bayesian Network model, a probabilistic graphical modelling approach, to capture the key factors and their interactions in Ethereum mainnet, the public Ethereum blockchain, focussing on the changes being introduced by Stateless Ethereum to estimate the health of the resulting Ethereum ecosystem. We use a mixture of empirical data and expert knowledge, where data are unavailable, to quantify the model. Based on the data and expert knowledge available to use at the time of modelling, the Ethereum ecosystem is expected to remain healthy following the introduction of Stateless Ethereum.

stat.AP↗

arXiv2022

On the intrinsic dimensionality of Covid-19 data: a global perspective

This paper aims to develop a global perspective of the complexity of the relationship between the standardised per-capita growth rate of Covid-19 cases, deaths, and the OxCGRT Covid-19 Stringency Index, a measure describing a country's stringency of lockdown policies. To achieve our goal, we use a heterogeneous intrinsic dimension estimator implemented as a Bayesian mixture model, called Hidalgo. We identify that the Covid-19 dataset may project onto two low-dimensional manifolds without significant information loss. The low dimensionality suggests strong dependency among the standardised growth rates of cases and deaths per capita and the OxCGRT Covid-19 Stringency Index for a country over 2020-2021. Given the low dimensional structure, it may be feasible to model observable Covid-19 dynamics with few parameters. Importantly, we identify spatial autocorrelation in the intrinsic dimension distribution worldwide. Moreover, we highlight that high-income countries are more likely to lie on low-dimensional manifolds, likely arising from aging populations, comorbidities, and increased per capita mortality burden from Covid-19. Finally, we temporally stratify the dataset to examine the intrinsic dimension at a more granular level throughout the Covid-19 pandemic.

stat.AP↗

arXiv2022

SSNbayes: An R package for Bayesian spatio-temporal modelling on stream networks

Spatio-temporal models are widely used in many research areas from ecology to epidemiology. However, most covariance functions describe spatial relationships based on Euclidean distance only. In this paper, we introduce the R package SSNbayes for fitting Bayesian spatio-temporal models and making predictions on branching stream networks. SSNbayes provides a linear regression framework with multiple options for incorporating spatial and temporal autocorrelation. Spatial dependence is captured using stream distance and flow connectivity while temporal autocorrelation is modelled using vector autoregression approaches. SSNbayes provides the functionality to make predictions across the whole network, compute exceedance probabilities and other probabilistic estimates such as the proportion of suitable habitat. We illustrate the functionality of the package using a stream temperature dataset collected in Idaho, USA.

stat.CO↗

arXiv2022

Bayesian spatio-temporal models for stream networks

Spatio-temporal models are widely used in many research areas including ecology. The recent proliferation of the use of in-situ sensors in streams and rivers supports space-time water quality modelling and monitoring in near real-time. A new family of spatio-temporal models is introduced. These models incorporate spatial dependence using stream distance while temporal autocorrelation is captured using vector autoregression approaches. Several variations of these novel models are proposed using a Bayesian framework. The results show that our proposed models perform well using spatio-temporal data collected from real stream networks, particularly in terms of out-of-sample RMSPE. This is illustrated considering a case study of water temperature data in the northwestern United States.

stat.ME↗

arXiv2021

A Data- and Task- Oriented Design Framework for Bivariate Communication of Uncertainty

The communication of uncertainty estimates, predictions and insights based on spatio-temporal models is important for decision-making as it impacts the utilisation and interpretation of information. Bivariate mapping is commonly used for communication of estimates and associated uncertainty; however, it is known that different visual qualities resulting from choics of symbols and consequent interaction between the display dimensions can lead to different interpretations and consequently affect resultant decisions. Characteristics of the data to be presented, such as spatial format, statistical level and continuousness, shape the range of available bivairate symbols. The subsequent utility of these bivariate symbols depends on their ability to achieve end-user's goals. In this paper we present a novel design framework, which, through consideration of both input data characteristics and potential operational tasks (as proxy to end-user goals), assists map designers in appropriate selection of bivariate symbols for the coincident presentation of spatio-temporal modelled data and associated uncertainty. The framework is showcased through application to a case study pertaining to sediment pollution in the Great Barrier Reef.

cs.HC↗

arXiv2021

Detection of cybersecurity attacks through analysis of web browsing activities using principal component analysis

Organizations such as government departments and financial institutions provide online service facilities accessible via an increasing number of internet connected devices which make their operational environment vulnerable to cyber attacks. Consequently, there is a need to have mechanisms in place to detect cyber security attacks in a timely manner. A variety of Network Intrusion Detection Systems (NIDS) have been proposed and can be categorized into signature-based NIDS and anomaly-based NIDS. The signature-based NIDS, which identify the misuse through scanning the activity signature against the list of known attack activities, are criticized for their inability to identify new attacks (never-before-seen attacks). Among anomaly-based NIDS, which declare a connection anomalous if it expresses deviation from a trained model, the unsupervised learning algorithms circumvent this issue since they have the ability to identify new attacks. In this study, we use an unsupervised learning algorithm based on principal component analysis to detect cyber attacks. In the training phase, our approach has the advantage of also identifying outliers in the training dataset. In the monitoring phase, our approach first identifies the affected dimensions and then calculates an anomaly score by aggregating across only those components that are affected by the anomalies. We explore the performance of the algorithm via simulations and through two applications, namely to the UNSW-NB15 dataset recently released by the Australian Centre for Cyber Security and to the well-known KDD'99 dataset. The algorithm is scalable to large datasets in both training and monitoring phases, and the results from both the simulated and real datasets show that the method has promise in detecting suspicious network activities.

cs.CR↗

arXiv2021

Understanding links between water-quality variables and nitrate concentration in freshwater streams using high-frequency sensor data

Real time monitoring using in situ sensors is becoming a common approach for measuring water quality within watersheds. High frequency measurements produce big data sets that present opportunities to conduct new analyses for improved understanding of water quality dynamics and more effective management of rivers and streams. Of primary importance is enhancing knowledge of the relationships between nitrate, one of the most reactive forms of inorganic nitrogen in the aquatic environment, and other water quality variables. We analysed high frequency water quality data from in situ sensors deployed in three sites from different watersheds and climate zones within the National Ecological Observatory Network, USA. We used generalised additive mixed models to explain the nonlinear relationships at each site between nitrate concentration and conductivity, turbidity, dissolved oxygen, water temperature, and elevation. Temporal auto correlation was modelled with an auto regressive moving average model and we examined the relative importance of the explanatory variables. Total deviance explained by the models was high for all sites. Although variable importance and the smooth regression parameters differed among sites, the models explaining the most variation in nitrate contained the same explanatory variables. This study demonstrates that building a model for nitrate using the same set of explanatory water quality variables is achievable, even for sites with vastly different environmental and climatic characteristics. Applying such models will assist managers to select cost effective water quality variables to monitor when the goals are to gain a spatially and temporally in depth understanding of nitrate dynamics and adapt management plans accordingly.

stat.AP↗

arXiv2021

Combined parameter and state inference with automatically calibrated ABC

State space models contain time-indexed parameters, termed states, as well as static parameters, simply termed parameters. The problem of inferring both static parameters as well as states simultaneously, based on time-indexed observations, is the subject of much recent literature. This problem is compounded once we consider models with intractable likelihoods. In these situations, some emerging approaches have incorporated existing likelihood-free techniques for static parameters, such as approximate Bayesian computation (ABC) into likelihood-based algorithms for combined inference of parameters and states. These emerging approaches currently require extensive manual calibration of a time-indexed tuning parameter: the acceptance threshold. We design an SMC$^2$ algorithm (Chopin et al., 2013, JRSS B) for likelihood-free approximation with automatically tuned thresholds. We prove consistency of the algorithm and discuss the proposed calibration. We demonstrate this algorithm's performance with three examples. We begin with two examples of state space models. The first example is a toy example, with an emission distribution that is a skew normal distribution. The second example is a stochastic volatility model involving an intractable stable distribution. The last example is the most challenging; it deals with an inhomogeneous Hawkes process.

stat.CO↗

arXiv2021

A Bayesian latent allocation model for clustering compositional data with application to the Great Barrier Reef

Relative abundance is a common metric to estimate the composition of species in ecological surveys reflecting patterns of commonness and rarity of biological assemblages. Measurements of coral reef compositions formed by four communities along Australia's Great Barrier Reef (GBR) gathered between 2012 and 2017 are the focus of this paper. We undertake the task of finding clusters of transect locations with similar community composition and investigate changes in clustering dynamics over time. During these years, an unprecedented sequence of extreme weather events (cyclones and coral bleaching) impacted the 58 surveyed locations. The dependence between constituent parts of a composition presents a challenge for existing multivariate clustering approaches. In this paper, we introduce a finite mixture of Dirichlet distributions with group-specific parameters, where cluster memberships are dictated by unobserved latent variables. The inference is carried in a Bayesian framework, where MCMC strategies are outlined to sample from the posterior model. Simulation studies are presented to illustrate the performance of the model in a controlled setting. The application of the model to the 2012 coral reef data reveals that clusters were spatially distributed in similar ways across reefs which indicates a potential influence of wave exposure at the origin of coral reef community composition. The number of clusters estimated by the model decreased from four in 2012 to two from 2014 until 2017. Posterior probabilities of transect allocations to the same cluster substantially increase through time showing a potential homogenization of community composition across the whole GBR. The Bayesian model highlights the diversity of coral reef community composition within a coral reef and rapid changes across large spatial scales that may contribute to undermining the future of the GBR's biodiversity.

stat.AP↗

arXiv2021

Spatio-temporal quantile regression analysis revealing more nuanced patterns of climate change: a study of long-term daily temperature in Australia

Climate change is commonly associated with an overall increase in mean temperature in a defined past time period. Many studies consider temperature trends at the global scale, but the literature is lacking in in-depth analysis of the temperature trends across Australia in recent decades. In addition to heterogeneity in mean and median values, daily Australia temperature data suffers from quasi-periodic heterogeneity in variance. However, this issue has barely been overlooked in climate research. A contribution of this article is that we propose a joint model of quantile regression and variability. By accounting appropriately for the heterogeneity in these types of data, our analysis reveals that daily maximum temperature is warming by 0.21 Celsius per decade and daily minimum temperature by 0.13 Celsius per decade. However, our modeling also shows nuanced patterns of climate change depends on location, season, and the percentiles of the temperature series over Australia.

stat.AP↗

arXiv2021

A Bayesian social platform for inclusive and evidence-based decision making

Against the backdrop of a social media reckoning, this paper seeks to demonstrate the potential of social tools to build virtuous behaviours online. We must assume that human behaviour is flawed, the truth can be elusive, and as communities we must commit to mechanisms to encourage virtuous social digital behaviours. Societies that use social platforms should be inclusive, responsive to evidence, limit punitive actions and allow productive discord and respectful disagreement. Social media success, we argue, is in the hypothesis. Documents are valuable to the degree that they are evidence in service of, or to challenge an idea for a purpose. We outline how a Bayesian social platform can facilitate virtuous behaviours to build evidence-based collective rationality. The chapter outlines the epistemic architecture of the platform's algorithms and user interface in conjunction with explicit community management to ensure psychological safety. The BetterBeliefs platform rewards users who demonstrate epistemically virtuous behaviours and exports evidence-based propositions for decision-making. A Bayesian social network can make virtuous ideas powerful.

cs.SI↗

arXiv2020

Assessing the accuracy of individual link with varying block sizes and cut-off values using MaCSim approach

Record linkage is the process of matching together records from different data sources that belong to the same entity. Record linkage is increasingly being used by many organizations including statistical, health, government etc. to link administrative, survey, and other files to create a robust file for more comprehensive analysis. Therefore, it becomes necessary to assess the ability of a linking method to achieve high accuracy or compare between methods with respect to accuracy. In this paper, we evaluate the accuracy of individual link using varying block sizes and different cut-off values by utilizing a Markov Chain based Monte Carlo simulation approach (MaCSim). MaCSim utilizes two linked files to create an agreement matrix. The agreement matrix is simulated to generate re-sampled versions of the agreement matrix. A defined linking method is used in each simulation to link the files and the accuracy of the linking method is assessed. The aim of this paper is to facilitate optimal choice of block size and cut-off value to achieve high accuracy in terms of minimizing average False Discovery Rate and False Negative Rate. The analyses have been performed using a synthetic dataset provided by the Australian Bureau of Statistics (ABS) and indicated promising results.

stat.CO↗

arXiv2020

A statistical machine learning approach for benchmarking in the presence of complex contextual factors and peer groups

The ability to compare between individuals or organisations fairly is important for the development of robust and meaningful quantitative benchmarks. To make fair comparisons, contextual factors must be taken into account, and comparisons should only be made between similar organisations such as peer groups. Previous benchmarking methods have used linear regression to adjust for contextual factors, however linear regression is known to be sub-optimal when nonlinear relationships exist between the comparative measure and covariates. In this paper we propose a random forest model for benchmarking that can adjust for these potential nonlinear relationships, and validate the approach in a case-study of high noise data. We provide new visualisations and numerical summaries of the fitted models and comparative measures to facilitate interpretation by both analysts and non-technical audiences. Comparisons can be made across the cohort or within peer groups, and bootstrapping provides a means of estimating uncertainty in both adjusted measures and rankings. We conclude that random forest models can facilitate fair comparisons between organisations for quantitative measures including in cases on complex contextual factor relationships, and that the models and outputs are readily interpreted by stakeholders.

stat.AP↗

arXiv2020

Peer groups for organisational learning: clustering with practical constraints

Peer-grouping is used in many sectors for organisational learning, policy implementation, and benchmarking. Clustering provides a statistical, data-driven method for constructing meaningful peer groups, but peer groups must be compatible with business constraints such as size and stability considerations. Additionally, statistical peer groups are constructed from many different variables, and can be difficult to understand, especially for non-statistical audiences. We developed methodology to apply business constraints to clustering solutions and allow the decision-maker to choose the balance between statistical goodness-of-fit and conformity to business constraints. Several tools were utilised to identify complex distinguishing features in peer groups, and a number of visualisations are developed to explain high-dimensional clusters for non-statistical audiences. In a case study where peer group size was required to be small ($\leq 100$ members), we applied constrained clustering to a noisy high-dimensional data-set over two subsequent years, ensuring that the clusters were sufficiently stable between years. Our approach not only satisfied clustering constraints on the test data, but maintained an almost monotonic negative relationship between goodness-of-fit and stability between subsequent years. We demonstrated in the context of the case study how distinguishing features between clusters can be communicated clearly to different stakeholders with substantial and limited statistical knowledge.

stat.AP↗

arXiv2020

Improved assessment of the accuracy of record linkage via an extended MaCSim approach

Record linkage is the process of bringing together the same entity from overlapping data sources while removing duplicates. Huge amounts of data are now being collected by public or private organizations as well as by researchers and individuals. Linking and analysing relevant information from this massive data reservoir can provide new insights into society. However, this increase in the amount of data may also increase the likelihood of incorrectly linked records among databases. It has become increasingly important to have effective and efficient methods for linking data from different sources. Therefore, it becomes necessary to assess the ability of a linking method to achieve high accuracy or to compare between methods with respect to accuracy. In this paper, we improve on a Markov Chain based Monte Carlo simulation approach (MaCSim) for assessing a linking method. MaCSim utilizes two linked files that have been previously linked on similar types of data to create an agreement matrix and then simulates the matrix using a proposed algorithm developed to generate re-sampled versions of the agreement matrix. A defined linking method is used in each simulation to link the files and the accuracy of the linking method is assessed. The improvement proposed here involves calculation of a similarity weight for every linking variable value for each record pair, which allows partial agreement of the linking variable values. A threshold is calculated for every linking variable based on adjustable parameter "tolerance" for that variable. To assess the accuracy of linking method, correctly linked proportions are investigated for each record. The extended MaCSim approach is illustrated using a synthetic dataset provided by the Australian Bureau of Statistics (ABS) based on realistic data settings. Test results show higher accuracy of the assessment of linkages.

stat.CO↗

arXiv2020

Assessing the accuracy of record linkages with Markov chain based Monte Carlo simulation approach

Record linkage is the process of finding matches and linking records from different data sources so that the linked records belong to the same entity. There is an increasing number of applications of record linkage in statistical, health, government and business organisations to link administrative, survey, population census and other files to create a complete set of information for more complete and comprehensive analysis. To make valid inferences using a linked file, it is increasingly becoming important to assess the linking method. It is also important to find techniques to improve the linking process to achieve higher accuracy. This motivates to develop a method for assessing linking process and help decide which linking method is likely to be more accurate for a linking task. This paper proposes a Markov Chain based Monte Carlo simulation approach, MaCSim for assessing a linking method and illustrates the utility of the approach using a realistic synthetic dataset received from the Australian Bureau of Statistics to avoid privacy issues associated with using real personal information. A linking method applied by MaCSim is also defined. To assess the defined linking method, correct re-link proportions for each record are calculated using our developed simulation approach. The accuracy is determined for a number of simulated datasets. The analyses indicated promising performance of the proposed method MaCSim of the assessment of accuracy of the linkages. The computational aspects of the methodology are also investigated to assess its feasibility for practical use.

stat.CO↗

arXiv2020

Correcting misclassification errors in crowdsourced ecological data: A Bayesian perspective

Many research domains use data elicited from "citizen scientists" when a direct measure of a process is expensive or infeasible. However, participants may report incorrect estimates or classifications due to their lack of skill. We demonstrate how Bayesian hierarchical models can be used to learn about latent variables of interest, while accounting for the participants' abilities. The model is described in the context of an ecological application that involves crowdsourced classifications of georeferenced coral-reef images from the Great Barrier Reef, Australia. The latent variable of interest is the proportion of coral cover, which is a common indicator of coral reef health. The participants' abilities are expressed in terms of sensitivity and specificity of a correctly classified set of points on the images. The model also incorporates a spatial component, which allows prediction of the latent variable in locations that have not been surveyed. We show that the model outperforms traditional weighted-regression approaches used to account for uncertainty in citizen science data. Our approach produces more accurate regression coefficients and provides a better characterization of the latent process of interest. This new method is implemented in the probabilistic programming language Stan and can be applied to a wide number of problems that rely on uncertain citizen science data.

stat.AP↗