We develop a learning rule that generalises the well known fading memory learning in the sense that the weights attached to the available time series data are not constant and are updated in light of the forecast error(s). The underlying idea is that confidence in the available data will be low...

We build a model in which asset prices are expectationally driven and agents forecast future prices hinging on a combination of fundamental value, trend and inertia. The model has a unique steady state and we investigate its stability. In particular the amount of behavioural heterogeneity in the...

We analyse the capacity of a range of different learning rules to describe actual human behaviour in the experiment on expectation formation in a cobweb model conducted by Hommes et al. (2000). We find indication of a relative superiority in terms of descriptive capacity of forms of generalised...