Mathematics
Te Tari Pāngarau me te Tatauranga
Department of Mathematics & Statistics

## Chuen Yen Hong

Studying for Doctor of Philosophy

Area of study:
Model averaging in meta-analysis

Supervisor: David Fletcher

### Thesis

Title: Focussed Model Averaging in GLMs

Supervisor: Tilman Davies

Previous Degree: BSc(Hons) (Otago)

In the frequentist framework, the model-averaged point estimate is a weighted mean of the estimates obtained from each model, the weight associated with a model often being based on a measure of its predictive ability, such as AIC, or of the probability that it is the true model, such as BIC. Several authors have proposed the use of focussed model averaging, in which the weights are tailored to the choice of a focus parameter. Thus, different choices of focus parameter might lead to different model weights. We present an approach to focussed model averaging using results for maximum likelihood estimation when the model is misspecified. In particular, we look at the special case of generalised linear models (GLMs).