The next meeting will be held at 5pm, Wednesday 15th of June in Room S1, 7 George Square. This will be the last meeting before we take a summer break, so it would be great to see as many of you there as we can!
We have two speakers:
Compositional data refers to multivariate data representing parts of a whole. They are intrinsically co-dependent positive amounts carrying only relative information, typically expressed in percentages or equivalent units (e.g. chemical or nutritional compositions, time allocated to different activities or behaviours, multiparty electoral data, investment portfolios and so on). These particularities, when ignored, have been shown to cause both technical and interpretability issues in data analysis such as singularity and multicollinearity in linear models, results dependent on the scale and size of the composition and spurious correlations. Mapping compositions onto the ordinary real space by log-ratio coordinates has been proved to be a suitable approach. In this talk I will give an overview and illustrate the use of the tools available in R to facilitate principled statistical analysis and modelling with data of compositional nature.
Mike will speak about his PhD research, which has considered snow cover and melt across Scotland. His work has revolved around the historic Snow Survey of Great Britain dataset (DOI: 10.1080/14702541.2014.900184) and has included point and gridded modelling of snow cover and melt. He’s used R as a general programming language, as well as for data analysis. Expect the talk to cover parallel processing and working with databases.
For any newcomers (you’re very welcome!), here’s a map of where we’ll be.
Meetings are open for all to attend (e.g. no University of Edinburgh card required). Hope to see you there!