Below we showcase several projects in which SIH has used time series analysis and modelling. See all projects.
Bayesian Updating for Childhood Obesity Grant Proposal
SIH supported a grant proposal by the Centre for Translational Data Science, by demonstrating the value of using Bayesian modelling when collecting and analysing longitudinal data on childhood obesity. We built cross-sectional Bayesian variable selection models to select important factors and models for predicting children’s BMI, mental health and sleep quality across multiple ages, for each child in the Longitudinal Study of Australian Children (LSAC) study. A vector-autoregressive model was then applied to visualise the unexplained variation in the preceding models. We constructed visualisations to demonstrate the importance of understanding uncertainty over the course of data collection, and the potential for using Bayesian adaptive trials during collection.
Identifying ram mating behaviour
Monitoring livestock has historically been labour intensive. The advent of on-animal sensors means this monitoring can be conducted remotely, continuously, and accurately. The ability to identify the precise time when sheep are mating using ram-mounted accelerometer data would unlock unprecedented information on the reproductive performance of these animals. We fit a classifier model to data from collar accelerometers labelled by videoing rams in the presence of ewes in oestrus. We then wrote code to detect change points in new acceleration data and to predict the occurence of mating events.
Video Tracking Predator-Prey Interactions in Fish.
By video-tracking the interaction between prey mosquitofish, Gambusia holbrooki, and their predator, jade perch, Scortum barcoo, under controlled conditions, we provide some of the first fine-scale characterisation of how prey adapt their behaviour according to their continuous assessment of risk based on both predator behaviour and angular distance to the predator’s mouth. When these predators were inactive and posed less of an immediate threat, prey were often found within the attack cone of the predator showing reductions in speed and acceleration, characteristic of predator-inspection behaviour. However, when predators became active, prey swam faster with greater acceleration and were closer together within the attack cone of predators. Most importantly, this study provides evidence that prey do not adopt a uniform response to the presence of a predator. Instead, we demonstrate that prey are capable of rapidly and dynamically updating their assessment of risk and showing fine-scale adjustments to their behaviour.
Paper: “Fine-scale behavioural adjustments of prey on a continuum of risk”. M.I.A. Kent, J.E. Herbert-Read, G.D. McDonald, A.J. Wood, A.J.W. Ward. Proceedings of the Royal Society B. 2019
1000-fold speedup in Dynamic Bayesian network model
A Bayesian network is a series of linear models fit to describe the relationships between different variables in a time series. If there are change points in how these variables are related, then the network is dynamic.
SIH helped the researcher by speeding up the R-package used to fit the dynamic bayesian network model by 1000x. The R-package is now available at https://github.com/FrankD/EDISON/tree/MultipleTimeSeries
Predicting Crime using a Spatial-Demographic Framework
Responding to domestic violence related assaults dominate much of the NSW Police’s resources. We try to understand the relationships that drive social-demographic change and cause the occurrence of crime using a complex modelling framework. The social-demographic-crime network and its inter-dependencies were modelled using a Bayesian vector autoregression model. We built a collaboration with BOCSAR, the crime database of all offences in NSW over the last 20 years, and sourced demographic data for multiple census years. The results of this study will help inform policy decision-making by government and police.
Clustering Light Sources: Scaling Up to the Whole Sky
The Murchison Widefield Array is a state-of-the-art telescope in Western Australia. Over the last four years, researchers have collected an exceptionally large time-series dataset on 300,000 bright objects in the sky, such as supernovae. Analysing the brightness of light sources over time requires matching each across pictures from different times and locations in the sky. The astrophysicists had built processing software, a database and a web app to analyse similar datasets, but had never tried to scale it to this size of dataset. An SIH engineer was able to debug and optimise the software involved, so that the data loading process took 8 hours instead of around 15 hours, and web app load times were reduced from multiple minutes to a few seconds. This enabled further research and analysis of this unique and enormous dataset.