Modeling data with structural and temporal correlation using lower level and higher level multilevel models

Novel imaging techniques are playing an increasingly important role in drug development, providing insight into the mechanism of action of new chemical entities. The data sets obtained by these methods can be large with complex inter-relationships, but the most appropriate statistical analysis for handling this data is often uncertain - precisely because of the exploratory nature of the way the data are collected. We present an example from a clinical trial using magnetic resonance imaging to assess changes in atherosclerotic plaques following treatment with a tool compound with established clinical benefit. We compared two specific approaches to handle the correlations due to physical location and repeated measurements: two-level and four-level multilevel models. The two methods identified similar structural variables, but higher level multilevel models had the advantage of explaining a greater proportion of variation, and the modeling assumptions appeared to be better satisfied.

An overview of domestic energy models and a new approach with agent based modelling

Government targets for CO2 reductions are being progressively tightened, the Climate Change Act set the UK target as an 80% reduction by 2050 on 1990 figures. The residential sector accounts for about 30% of emissions. This paper discusses current modelling techniques in the residential sector: principally top-down and bottom-up. Top-down models work on a macro-economic basis and can be used to consider large scale economic changes; bottom-up models are detail rich to model technological changes. Bottom-up models demonstrate what is technically possible. However, there are differences between the technical potential and what is likely given the limited economic rationality of the typical householder. This paper recommends research to better understand individuals’ behaviour. Such research needs to include actual choices, stated preferences and opinion research to allow a detailed understanding of the individual end user. This increased understanding can then be used in an agent based model (ABM). In an ABM, agents are used to model real world actors and can be given a rule set intended to emulate the actions and behaviours of real people. This can help in understanding how new technologies diffuse. In this way a degree of micro-economic realism can be added to domestic carbon modelling. Such a model should then be of use for both forward projections of CO2 and to analyse the cost effectiveness of various policy measures.

Liquid crystalline elastomers: the relationship between macroscopic behaviour and the level of backbone anisotropy

Nematic monodomain liquid crystalline elastomers have been prepared through in situ cross-linking of an acrylate based side-chain liquid crystalline polymer in a magnetic field. At the nematic–isotropic transition, the sample is found to undergo an anisotropic shape change. There is found to be an increase in dimensions perpendicular — and a decrease parallel — to the director, this is consistent with alignment of the polymer backbone parallel to the direction of mesogen alignment in the nematic state. From a quantitative investigation of this behaviour, we estimate the level of backbone anisotropy for the elastomer. As second measure of the backbone anisotropy, the monodomain sample was physically extended. We have investigated, in particular, the situation where a monodomain sample is deformed with the angle between the director and the extension direction approaching 90°. The behaviour on extension of these acrylate samples is related to alternative theoretical interpretations and the backbone anisotropy determined. Comparison of the chain anisotropy derived from these two approaches and the value obtained from previous small-angle neutron scattering measurements on deuterium labelled mixtures of the same polymer shows that some level of chain anisotropy is retained in the isotropic or more strictly weakly paranematic state of the elastomer. The origin and implications of this behaviour are discussed.