Uncategorized

Spring term 2023 SE@K Thursday lunches

Spring term starts with the second half of Python training by Lena :-). Plan for the term in terms of sessions (and cake) below!

  1.  19th of January     Session : Python training II – Lena; Cake : Lena & Daniel
  2.  26th of January     Session : Daniel’s grant; Cake : Diana
  3.  2nd of February    Session : HMM paper – Diana; Cake : Eleni
  4.  9th of February     Session : HMM paper – Daniel; Cake : Oscar
  5.  16th of February   Session : HMM paper – Takis (guest lecture); Cake : Alex
  6.  23rd of February  Session : Variable selection/Hypothesis testing – Alex; Cake : James
  7.  2nd of March        Session : HMM – Fred (guest lecture); Cake : Fabian
  8.  9th of March         Session : Parameter redundancy – Daniel’s take; Cake : Tommy
  9. 16th of March         Session Swedish register data – Bruno&Eleni ; Cake : Milly
  10. 23th of March         Session : Swedish register data – Bruno&Eleni; Cake : Daniel
  11.  30th of March        Session : Spatio-temporal models – Oscar; Cake : Diana
  12.  6th of April             Session : eDNA data – Alex; Cake :  Eleni
Standard
Uncategorized

Se@k weekly lunches

The SE@K group are meeting weekly for research lunch and socialising 🙂

Autumn term

First week was about introductions and a tasty lunch at Dolce Vita!

Second week was about staff talking about their research interests, with Diana talking to the group about parameter redundancy.

Third, fourth & fifth week were about research students talking about their research, with Tommy using randomised response techniques to estimate the proportion of people in the group who like/liked their supervisor (!), Milly talking about distance sampling, Ioannis talking about Bayesian non-parametrics and ABC and Lena talking about predator-prey models!

The term went on with learning about Gaussian processes from Alex, about PDEs from Eduard and about Python from Lena!

 

 

 

Standard
News, Papers, Publications

New paper in Environmetrics by Alex

A vector of point processes for modeling interactions between and within species using capture-recapture data

Alex Diana, Eleni Matechou, Jim E. Griffin, Yadvendradev Jhala, Qamar Qureshi

Abstract

Capture-recapture (CR) data and corresponding models have been used extensively to estimate the size of wildlife populations when detection probability is less than 1. When the locations of traps or cameras used to capture or detect individuals are known, spatially-explicit CR models are used to infer the spatial pattern of the individual locations and population density. Individual locations, referred to as activity centers (ACs), are defined as the locations around which the individuals move. These ACs are typically assumed to be independent, and their spatial pattern is modeled using homogeneous Poisson processes. However, this assumption is often unrealistic, since individuals can interact with each other, either within a species or between different species. In this article, we consider a vector of point processes from the general class of interaction point processes and develop a model for CR data that can account for interactions, in particular repulsions, between and within multiple species. Interaction point processes present a challenge from an inferential perspective because of the intractability of the normalizing constant of the likelihood function, and hence standard Markov chain Monte Carlo procedures to perform Bayesian inference cannot be applied. Therefore, we adopt an inference procedure based on the Monte Carlo Metropolis Hastings algorithm, which scales well when modeling more than one species. Finally, we adopt an inference method for jointly sampling the latent ACs and the population size based on birth and death processes. This approach also allows us to adaptively tune the proposal distribution of new points, which leads to better mixing especially in the case of non-uniformly distributed traps. We apply the model to a CR data-set on leopards and tigers collected at the Corbett Tiger Reserve in India. Our findings suggest that between species repulsion is stronger than within species, while tiger population density is higher than leopard population density at the park.

Standard
PhD projects

Fully funded studentship in collaboration with Butterfly Conservation

PROJECT TITLE:

Developing novel spatio-temporal models for large citizen science data sets

Supervisors: This is a joint project between the University of Kent and Butterfly Conservation and the PhD student will be supervised by a team with expertise in Statistics, Statistical Ecology, Citizen Science, and Butterfly Monitoring

University of Kent: Dr Eleni Matechou, Dr Diana Cole, Prof  Byron Morgan

Butterfly Conservation: Dr Emily Dennis, Dr Richard Fox

Application Process

APPLY NOW

Deadline for applications: midnight 2nd of July 2023. 

Interviews will take place on the 18th of July 2023.

Applicants should follow the University of Kent’s online application process.

Please create an account and add your personal details as requested. Subsequently, you need to select your starting date (September 2023) and write your personal statement (see below). Choose “Other” for source of funding and “Definite” for funding. Add details of your qualifications, and then in the Research Information Tab, write “Dr Eleni Matechou” under supervisor and the title of the project (“Developing novel spatio-temporal models for large citizen science data sets”) as the research topic. You do not need to add a research proposal.

As part of the process, you need to provide the following:

o   details of your qualifications;

o   two academic references;

o  a personal statement

The statement must be maximum 500 words detailing (1) your reason for applying for a doctoral studentship (i.e, why do you want to pursue doctoral studies) and (2) your fit with the proposed project (how your educational/professional/personal background has prepared you well to undertake research in this topic).

Please email Dr Eleni Matechou (e.matechou@kent.ac.uk) if you are interested in applying for the project or have any questions about the project or the application process.

Person specification

We seek a candidate with a strong quantitative background, for example an MSc in Statistics or an MSc with high statistics content. Experience coding in R, or similar, is essential. An interest in conservation and ecology is advantageous.

The University of Kent requires all non-native speakers of English to reach a minimum standard of proficiency in written and spoken English before beginning a postgraduate degree. For more information on English language requirements, please visit this page.

Training

The student will develop a strong, highly transferable skillset in statistical modelling and analysis using modern statistical and computational techniques applied to large unstructured data sets, with spatial and temporal replications. The student will benefit from interactions with conservation professionals at Butterfly Conservation, including opportunities to undertake fieldwork, to better understand the data collection processes and focal taxa of the project, as well as data use for conservation delivery and policy.

Scientific background

At a time of biodiversity loss, including widely reported insect declines, monitoring changes in species’ populations and distributions is vital. To that effect, there is an ongoing growth of unstructured citizen science data, where species can be recorded by anyone, from any time and place. The full potential of such sources of increasingly ‘big data’ for biodiversity monitoring has not yet been fully realized.

Analysing citizen science data of this nature presents unique challenges relating to their vast quantity but also associated sampling biases. Using cutting edge modelling, this project will maximise the potential of these valuable datasets, with application to data for butterflies and moths, to enhance our understanding of species’ phenology (flight periods), distribution and range dynamics.

Butterflies and moths respond quickly to habitat and climatic change, and hence are valuable biodiversity indicators. In the UK, millions of species’ occurrence records for Lepidoptera have been gathered by two large citizen science recording schemes. This project will develop new models for these datasets, with the aim to help inform future conservation delivery and policy and better understand the drivers of species’ change.

 

Research methodology

The student will develop new Bayesian statistical models and associated efficient algorithms and apply them to large spatio-temporal citizen science data for single and multiple species. The research will involve development of

  • state-of-the-art variable selection techniques to better describe drivers of species’ range and distribution change through suitable spatial and environmental covariates.
  • novel approaches to density estimation, building on Bayesian nonparametrics and related approaches, for modelling species’ phenology from citizen science data to identify drivers of change.
  • assessment approaches to determine reliability of inference for rare and/or under-recorded species from available data.

Research excellence

The student will join the thriving Statistical Ecology @ Kent research group, being supervised by leading researchers in statistics and statistical ecology. They will also be members of the UK-wide National Centre for Statistical Ecology. They will attend London Taught Course Centre training, NCSE seminars, and SE@K specialist training and they will present research results at a range of appropriate national and international conferences. There will be ample opportunity for independent development, with the student gaining transferable knowledge of modern data science and statistics.

 Scholarship Information

VC scholars will receive the following:

  • Annual stipend at UKRI rates (£17,668 in 2022/23);
  • Annual tuition fees at Home rates (£4,596 in 2022/23)

o   2023/24 rates to be announced.

Relevant literature

Diana, A., Dennis, E. B., Matechou, E., Morgan, B. J. T. (2022) Fast Bayesian inference for large occupancy datasets. Biometrics, . ISSN 0006-341X. (In press) (KAR id:98286)

Dennis, E.B., Morgan, B.J.T., Freeman, S.N., Ridout, M.S., Brereton, T.M., Fox, R., Powney, G.D., Roy, D.B. (2017) Efficient occupancy model-fitting for extensive citizen-science data. PLoS ONE 12(3): e0174433. https://doi.org/10.1371/journal.pone.0174433

Diana, A., Matechou, E., Griffin, J., Arnold, T., Tenan, S. & Volponi, S. (2022) A general modeling framework for open wildlife populations based on the Polya tree prior. Biometrics, 00, 1– 13. https://doi.org/10.1111/biom.13756

Griffin, J. E., Matechou, E., Buxton, A. S., Bormpoudakis, D., & Griffiths, R. A. (2020). Modelling environmental DNA data; Bayesian variable selection accounting for false positive and false negative errors. Journal of the Royal Statistical Society: Series C (Applied Statistics), 69(2), 377-392.

Dennis, E.B., Morgan, B,J.T, Freeman, S.N., Brereton, T.M. & Roy, D.B. (2016). A generalized abundance index for seasonal invertebrates. Biometrics, 71, 1305-1314.

Dennis, E.B., Brereton, T.M., Morgan, B.J.T., Fox, R., Shortall, C.R., Prescott, T. & Foster, S. (2019). Trends and indicators for quantifying moth abundance and occupancy in Scotland. Journal of Insect Conservation, 23, 369-380.

Standard
News, PhD projects

Migration and Movement SRT project

Dr Eleni Matechou and Dr Bruno Santos have been awarded funding from the Migration and Movement SRT for their project on Modelling Human Population Registers.

PROJECT TITLE: MODELLING OF HUMAN POPULATION REGISTERS

Supervisors: The PhD student will be supervised by an interdisciplinary team across four international institutions, with expertise in

(1)  Statistics (Dr Bruno Santos) and Statistical Ecology (Dr Eleni Matechou), University of Kent, Canterbury, England

(2) Demography (Dr Eleonora Mussino), University of Stockholm, Sweden

(3)  Bayesian Inference/Statistical Ecology (Professor Ruth King), University of Edinburgh, Scotland

(4) Ecology/Statistical Ecology (Dr Blanca Sarzo), University of Valencia, Spain

Research background: Monitoring the size and characteristics of human populations using official censuses is a lengthy and costly process. As a result, in recent years, there has been an increased focus on using a statistical framework instead, referred to as multiple systems estimation (MSE), to infer population size of specific groups using opportunistic registers and incomplete lists. Examples include estimating the number of drug users in a city [1] and the number of victims of human trafficking [2]. MSE builds on well-established statistical theory and models, and can be employed using existing software [3]. However, MSE does not follow the same individuals over time to learn from their past experiences or behaviours and hence cannot account for individual heterogeneity in the probability of being observed in one or more registers, for dependence between individuals or to identify the factors behind temporary emigration.

These incomplete and imperfect human population registers are similar to ecological data, referred to as capture-recapture (CR) data, collected on wildlife populations, such as birds and mammals [4]. The corresponding ecological CR models have the same aims as MSE, namely estimating population size and monitoring population characteristics, but rely on completely different modelling approaches, with a strong focus on modelling individual time series [5]. However, CR models are computationally more demanding than MSE, and as a result do not scale well to large populations [6].

Project aims: The project aims to bring together MSE and CR modelling approaches and provide a general and unifying modelling framework for human population registers. The new models to be developed will overcome the shortcomings of the existing approaches, and hence will be applicable to high-dimensional data sets typically observed in human populations, and increasingly in wildlife populations, whilst at the same time modelling individual time-series data.

The models will be applied to data from several countries that routinely collect register data, such Sweden, Norway and Italy. We will obtain estimates of population size for different sub-populations, such as the migrant population each year, as well as of the probability of individuals of certain characteristics, such as age, sex etc., to appear in each register, e.g. employment. These estimates will then be compared to currently used metrics by each country to quantify the effect of over-coverage, which is the bias introduced by considering individuals who are not part of the population in demographic rates, such as mortality. The developed models and corresponding code will be made freely available and the methods will be disseminated through close collaboration with demographers in the different countries and corresponding workshops, as appropriate.

References

[1] King, R., Bird, S. M., Overstall, A. M., Hay, G., & Hutchinson, S. J. (2014). Estimating prevalence of injecting drug users and associated heroin‐related death rates in England by using regional data and incorporating prior information. Journal of the Royal Statistical Society: Series A (Statistics in Society), 177(1), 209-236. [2] Zhang, S. X., & Larsen, J. J. (2021). Estimating the size of the human trafficking problem: MSE and other strategies. Crime & Delinquency, 67(13-14), 2169-2187. [3] Overstall, A., & King, R. (2014). conting: An R package for Bayesian analysis of complete and incomplete contingency tables. Journal of Statistical Software, 58(7), 1-27. [4] McCrea, R. S., & Morgan, B. J. (2014). Analysis of capture-recapture data. CRC Press. [5] Matechou, E., & Argiento, R. (2022). Capture-recapture models with heterogeneous temporary emigration. Journal of the American Statistical Association, (just-accepted), 1-32. [6] King, R., Sarzo, B., & Elvira, V. (2022). Large Data and (Not Even Very) Complex Ecological Models: When Worlds Collide. arXiv preprint arXiv:2205.07261.

Research excellence: The student will join the thriving Statistical Ecology @ Kent research group, and the Migration and Movement Signature Research Theme, being supervised by leading researchers in demography, statistics and statistical ecology. They will also be members of the UK-wide National Centre for Statistical Ecology. They will attend London Taught Course Centre training, NCSE seminars, and SE@K specialist training and they will present research results at a range of appropriate national and international conferences. There will be ample opportunity for independent development, with the student gaining transferable knowledge of modern data science and statistics.

Please email Dr Eleni Matechou (e.matechou@kent.ac.uk) if you are interested in applying for the project or have any questions about the project or the application process.

Application Process

Applicants should follow the University of Kent’s online application process.

Please create an account and add your personal details as requested. Subsequently, you need to select your starting date (September 2023) and write your personal statement (see below). Choose “Other” for source of funding and “Definite” for funding. Add details of your qualifications, and then in the Research Information Tab, write “Dr Eleni Matechou” under supervisor and the title of the project (“Modelling of human population registers“) as the research topic. You do not need to add a research proposal.

As part of the process, you need to provide the following:

o   details of your qualifications;

o   two academic references;

o  a personal statement

The statement must be maximum 500 words detailing (1) your reason for applying for a doctoral studentship (i.e, why do you want to pursue doctoral studies) and (2) your fit with the proposed project (how your educational/professional/personal background has prepared you well to undertake research in this topic).

Entry Requirements

We seek a candidate with a strong quantitative background, for example an MSc in Statistics or an MSc with high statistics content, or a background in demographic modelling. Experience coding in R, or similar, is essential.

The University of Kent requires all non-native speakers of English to reach a minimum standard of proficiency in written and spoken English before beginning a postgraduate degree. For more information on English language requirements, please visit this page.

About the Migration and Movement Signature Research Theme (SRT)

The Migration and Movement SRT is a vibrant community of over 100 scholars and practitioner academics working in the field of migration and movement, led by Prof. David Herd (English), Dr Bahriye Kemal (English), Dr Amanda Klekowski von Koppenfels (Politics), Dr Margherita Laera (Drama), Dr Tom Parkinson (Higher Education) and Dr Sweta Rajan-Rankin (Sociology). Members bring a number of interdisciplinary perspectives, from International Law to Urban Planning, Medicine, Health, Pharmacy, Statistics, Politics and Entrepreneurship. The SRT aims to expand our understanding of migration beyond the movement of people to include the migration of pathogens, remittances, technologies, cultures, scriptures, coffee, drugs, medicines, labour, and ideas.

The Migration and Movement SRT was launched in September 2021, at a historical moment when intersecting crises of movement were (and still are) taking place. As the COVID-19 pandemic was spreading, governments were enforcing lockdowns and movement restrictions, yet the UNHCR reported forced displacement at upwards of 75 million people. The year 2022 opened with new refugee crises in Ukraine and Ethiopia, while climate catastrophes and other military conflicts reaped havoc across the planet, notably in Pakistan. People facing hardship will continue to seek sustainable existences elsewhere and will be compelled to move, yet governments in the global north are preparing to render the seeking of asylum more difficult or even illegal.

As a public-facing, civic-oriented research group, we believe universities – the University of Kent  in particular – can play a key role as leaders of intellectual debates and as creators of advocacy and evidence-based research around migration, its contexts, histories and benefits. We are committed to learning from the work of decolonial and postcolonial colleagues, and from the perspectives of persons with lived experience. Reflexive learning and intersectional empathy underpin every aspect of our ethics. We are keen to foster the next generation of scholars working on issues of migration and movement, broadly conceived, and welcome applications from scholars currently living in all parts of the world. For more information on the Migration and Movement SRT, please visit: https://research.kent.ac.uk/signature-themes/migration-and-movement/

 Scholarship Information

Signature Research Theme scholars will receive the following:

  • Annual stipend at UKRI rates (£17,668 in 2022/23);
  • Annual tuition fees at Home rates (£4,596 in 2022/23)

o   2023/24 rates to be announced.

Home and International candidates are eligible to apply. The Abdulrazak Gurnah Doctoral Scholarship will cover International fees, while the Migration and Movement Doctoral Scholarship will cover Home fees only, including for those with UK settled or pre-settled status.

 

Selection Process

Supervisors will carry out the initial eligibility checks and shortlisting. All shortlisted candidates will then be ratified at Division level where a further selection process might occur. Four final shortlisted candidates per project will be submitted to the SRT Studentship Selection Panel.

Shortlisted candidates will be invited for an interview taking place the week commencing 27 February 2023.

Deadline

The deadline for  applications is midnight on 3 February.

Standard
News, Papers

JASA paper by Eleni and Raffaele

The paper, titled “Capture-recapture models with heterogeneous temporary emigration”, first published online on the 14th of September 2022, develops a novel modelling framework for capture-recapture data without relying on the assumption of permanent emigration. The model is built within a Bayesian non-parametric & changepoint process framework, and is demonstrated on data on salmon anglers in Norway.

 

 

Standard
Uncategorized

rGAI: An R package for fitting the generalized abundance index to seasonal count data

Emily Dennis, Calliste Fagard-Jenkin and Byron Morgan have created an R package for fitting the generalized abundance index to seasonal count data. The work has been published in Ecology and Evolution in the paper “rGAI: An R package for fitting the generalized abundance index to seasonal count data”.

The paper can be found at: https://onlinelibrary.wiley.com/doi/full/10.1002/ece3.9200

The R package is available at: https://github.com/calliste-fagard-jenkin/rGAI

 

Standard