Luke Mosley

Luke Mosley

Luke Mosley



I am currently based in Lancaster doing a PhD in Model Selection for High-Dimensional Temporal Disaggregation in Official Statistics. This work is in joint collaboration with the STOR-i Centre of Doctoral Training at Lancaster University and the Office for National Statistics. Before this, I completed a Masters in Statistics and Operation Research with STOR-i CDT gaining an MRes with distinction. My undergraduate degree was completed at Warwick University gaining a BSc(Hons) in Mathematics and Statistics. My research interests lie in the crossover between high-dimensional statistics and time series analysis, in particular, how we can fuse statistical learning concepts into the econometrics literature to help disaggregate official time series output.

Alongside my PhD I am a Graduate Teaching Assistant at Lancaster University teaching a range of modules including 1st year Maths modules such as Calculus, Probability and Linear Algebra, and 4th year Masters modules including Statistical Learning. Further to this, I have experienced teaching Maths and English to a much younger age group at Explore Learning as a part-time tutor there during my time at Warwick.

Away from studies, my two main passions are sport and music. Football is the main sport I have played for the majority of my life, currently representing the Graduate College team at Lancaster. I am also a keen cricketer being lucky enough to represent both Wigan District and Greater Manchester West during secondary school. In my spare time, I like to challenge myself to golf and snooker, both being frustratingly difficult. I am a big fan of rock and roll and love going to gigs and festivals.

Personal Information

  • NameLuke Mosley
  • Age23 Years
  • ResidenceLancaster, UK
  • OccupationPhD Student
  • UniversityLancaster


  • PhD, Time Series Analysis

    STOR-i CDT, Lancaster University and The Office for National Statistics

    Working with the Office for National Statistics to develop novel methods for combining insight from panel data, and higher-resolution observational time-series, with the goal to meet the ever-increasing user demand for improved and more detailed statistics.

    Present Oct 2019
  • MRes, Statistics and Operational research

    STOR-i CDT, Lancaster University

    Undertaking a Masters of Research in statistics and operational research, giving me an overview of thriving research areas and an opportunity to develop a formal research proposal for my PhD.

    Sept 2019 Oct 2018
  • BSc (Hons), Mathematics and Statistics

    Warwick University

    Provided me with a thorough grounding in both theoretical and practical aspects of modern statistics.

    July 2018 Oct 2015

Work Experience

  • Graduate Teaching Assistant

    Lancaster University

    Workshop tutor for the MATH100 series including: Calculus, Further Calculus, Probability, Statistics and Linear Algebra. Also, the workshop tutor for CFAS420 - Statistical Learning.

    Present Oct 2019
  • Temp leasehold administrator

    Onward Homes Ltd, Bolton

    My role involved carrying out administration duties for leasehold departments and going through a number of confidential and sensitive leaseholder files; scanning needed documents and destroying unneeded ones.

    Sep 2018 July 2018
  • Tutor

    Explore Learning, Leamington Spa

    My goal was to transfer a true passion for education, trying to motivate and inspire children from 4 to 14 to become confident learners in English and Mathematics.

    Feb 2018 Feb 2017
  • Waiter and bar man

    Tavernfayre, Bolton

    Maximise customer satisfaction by being energetic, quick to respond to problems and creating a welcoming atmosphere for customers.

    Jan 2017 Jan 2014



What is STOR-i?

The STOR-i Centre for Doctoral Training, a joint venture between the departments of Mathematics and Statistics and Management Science, offers a four-year PhD programme in Statistics and Operational Research (STOR) developed and delivered with industrial partners. STOR-i was established in 2010 as an EPSRC Centre for Doctoral Training. The Centre has developed an international reputation for the quality of its research. The multi-million pound award includes substantial investment from EPSRC, Lancaster University and several key industrial partners.

The programme

The training programme spans four years, consisting of a foundation year resulting in the award of a Masters of Research (MRes), followed by a three year period of study leading to a PhD. The first year includes taught courses, projects and group activities providing a grounding in Statistics and Operational Research, an overview of thriving research areas, and an opportunity to develop a formal research proposal for a PhD. In years 2-4 involves the completion of a PhD project, encountering real-life commercial challenges, developing leading-edge STOR research and making a real impact on major industrial and scientific applications.

To visit the STOR-i website, please click here.

Smiley face Smiley face Smiley face



My PhD project is titled: "Information fusion for non-homogeneous panel and time series data." This project is in collaboration with the STOR-i institute at Lancaster University and the Office for National Statistics (ONS). My academic supervisors at Lancaster are Professor Idris Eckley and Dr Alex Gibberd. My industrial supervisor at ONS is Dr Hannah Finselbach.

The ONS are transforming to put administrative and alternative data sources at the core of their statistics. Official statistics have traditionally been reliant on sample surveys and questionnaires, however, in this rapidly evolving economy, response rates of these surveys are falling. Moreover, there exists a concern of not making full use of new data sources and the continuously expanding volume of information that is now available. Today, information is being gathered in a countless number of ways, from satellite and sensory data, to social network and transactional data. There is certainly an opportunity to remodel the 20th century survey-centric way to a 21st century combination of structured survey data, with administrative and unstructured alternative digital data sources.

My PhD project is to assist the ONS with this transformation, by developing novel methods for combining insight from the alternative (possibly dynamic) information recorded at a different periodicity and reliability, with traditional surveys, in order to meet the ever-increasing demand for improved and more detailed statistics.

The time series data coming from traditional methods are typically recorded at a lower frequency (e.g. annually) and while accurate, and well callibrated, they are very expensive to run and take a long time to feed-back information. By additionally using administrative datasets and alternative data streams such as web-scarped data, we can potentially increase both the frequency and the accuracy at which official statistics are produced.

Understanding how and which of the vast collection of high frequency alternative indicator series are relevent to producing a particular statistic of interest is the predominant area of work I am considering at the start of the PhD. Given that only a few of the indictaor series are likely to be relevant, I explore incorporating sparse modelling techniques such as LASSO regularization in the econometrics time series literature. Beyond this I wish to look into theoretical properties of data revisions and also spatial high-frequency data by considering extensions to vector autoregressive (VAR) models.


The MRes component of the STOR-i programme includes taught courses, projects and group activities providing me with a grounding in statistics and operational research, an overview of thriving research areas, and an opportunity to develop a formal research proposal for my PhD. Below is a list of project reports I have completed this year. Also, check out my blog to read about some of the research areas I have been exposed to.

16/04/2019: Using adaptive random search for simulation optimisation - POSTER

31/03/2019: Modelling the demand of healthcare systems using infinite-server queueing models

24/02/2019: Detecting multiple changes in variance of univariate time series data

03/02/2019: Continuous-Time Markov chains and Discrete-Time Markov decision processes

03/12/2018: Branch and Bound algorithm to solve the Knapsack problem


Let's Talk