Estimation and Forecasting of Locally Stationary Processes

This paper develops a state space framework for the statistical analysis of a class of locally stationary processes. The proposed Kalman filter approach provides a numerically efficient methodology for estimating and predicting locally stationary models and allows for the handling of missing values. It provides both exact and approximate maximum likelihood estimates. Furthermore, as suggested by the Monte Carlo simulations reported in this work, the performance of the proposed methodology is very good, even for relatively small sample sizes. Copyright © 2011 John Wiley & Sons, Ltd.     

Capturing complex, non-linear team behaviours during competitive football performance

This study investigated changes in the complexity (magnitude and structure of variability) of the collective behaviours of association football teams during competitive performance. Raw positional data from an entire competitive match between two professional teams were obtained with the ProZone® tracking system. Five compound positional variables were used to investigate the collective patterns of performance of each team including: surface area, stretch index, team length, team width, and geometrical centre. Analyses involve the coefficient of variation (%CV) and approximate entropy (ApEn), as well as the linear association between both parameters. Collective measures successfully captured the idiosyncratic behaviours of each team and their variations across the six time periods of the match. Key events such as goals scored and game breaks (such as half time and full time) seemed to influence the collective patterns of performance. While ApEn values significantly decreased during each half, the %CV increased. Teams seem to become more regular and predictable, but with increased magnitudes of variation in their organisational shape over the natural course of a match.     

Modelling conservation in the Amazon basin

Expansion of the cattle and soy industries in the Amazon basin has increased deforestation rates and will soon push all-weather highways into the region's core. In the face of this growing pressure, a comprehensive conservation strategy for the Amazon basin should protect its watersheds, the full range of species and ecosystem diversity, and the stability of regional climates. Here we report that protected areas in the Amazon basin--the central feature of prevailing conservation approaches--are an important but insufficient component of this strategy, based on policy-sensitive simulations of future deforestation. By 2050, current trends in agricultural expansion will eliminate a total of 40% of Amazon forests, including at least two-thirds of the forest cover of six major watersheds and 12 ecoregions, releasing 32 +/- 8 Pg of carbon to the atmosphere. One-quarter of the 382 mammalian species examined will lose more than 40% of the forest within their Amazon ranges. Although an expanded and enforced network of protected areas could avoid as much as one-third of this projected forest loss, conservation on private lands is also essential. Expanding market pressures for sound land management and prevention of forest clearing on lands unsuitable for agriculture are critical ingredients of a strategy for comprehensive conservation.     

Experience in combining subjective and quantitative forecasts of open market rates

This paper presents a method of combining subjective information from open-market operators with results from a time-series forecasting model. Empirical results of forecasts for interest rates of bank reserves are presented.