Publications

We study learning dynamics in a prototypical representative-agent forward-looking model in which agents’ beliefs are updated using linear learning algorithms. We show that learning in this model can generate long memory endogenously, without any persistence in the exogenous shocks, depending on the weights agents place on past observations when they update their beliefs, and on the magnitude of the feedback from expectations to the endogenous variable. This is distinctly different from the case of rational expectations, where the memory of the endogenous variable is determined exogenously.

In this paper, I consider how to improve forecasting techniques when the underlying data generating process is subject to stochastic shifts. I show how the improvement relates to the forecasting horizon.

This article considers the issue of estimation and inference in models where agents' forward-looking expectations are formed via adaptive learning. We show that nonstandard dynamics result and invalidate standard techniques of inference. We provide valid methods and assess them by means of a three-equation new Keynesian macro model.

This papers evaluates the risk premium associated with the price of crude oil. The risk premium is defined as the difference between actual pricing and pricing based on historical demand and supply conditions

This article considers techniques for forecasting in a multivariate framework when the forecast horizon of interest varies and the economy is subject to breaks or change. This is mostly an empirical paper that assess how best to forecast the GDP of South Africa. 

This paper reviews existing results and provides a framework for understanding the interplay between the forecast horizon and the forecasting techniques.

We consider the optimal methods for forecasting as a function of the horizon at which it is wished to obtain a forecast.