Glenna Nightingale

Chancellor's Research Fellow

Biography

Background

I graduated from St Andrews University in 2013 with a PhD in Statistics.  My research focus was on spatial point process models in a Bayesian context. Since then, I have gained experience in modelling time to event data (in a Bayesian context), constructing R packages, analysing longitudinal data and spatiotemporal data, construction of complex survey weights, factor analysis and machine learning.   I possess broad skills covering both Bayesian and Frequentist statistical modelling, and have developed custom models that are perfectly suited to the task at hand rather than using off-the-shelf models. 

I also collaborate with other Statisticians and Epidemiologists– for example on projects that involve Area Interaction Point Process Models,  Log Gaussian Cox Processes,  time series modelling using ARIMA, HMM, and GLMs and COVID-19 surveillance. 

My hobbies include playing piano and building R Shiny apps for data visualization RSS Fellow in April 2018, Chancellor's Fellow (University of Edinburgh), 2021

http://www.glennanightingale.com/

 

Areas of interest for supervision

PhD Student Gosaye Kaba

Research summary

My research interests span constructing point process models in a Bayesian context to developing quantitative methods for evaluating complex public health interventions.

Current research interests

At present I am working on a project which evaluates the 20mph speed limit policy in the City of Edinburgh.

Past research interests

I have gained experience in modelling time to event data (in a Bayesian context), constructing R packages, analysing longitudinal data and spatiotemporal data, construction of complex survey weights, factor analysis and machine learning. I possess broad skills covering both Bayesian and Frequentist statistical modelling, and have developed custom models that are perfectly suited to the task at hand rather than using off-the-shelf models. I also collaborate with other Statisticians – for example on projects that involve Area Interaction Point Process Models, Log Gaussian Cox Processes, and time series modelling using ARIMA, HMM, and GLMs.
This article was published on