Speakers in several of the sessions were invited by the session organizers to prepare written versions of their presentations to appear in a refereed special issue. The papers here are those of some of the speakers who accepted the invitation, while some will have appeared in other issues of Environmetrics.
The ยฎrst papers here were presented at the Malaysian conference. The paper by Jacobsen et al. considers modelling the eects of solar radiation on the oxygen content of a small creek. Several models of oxygen level are presented and compared using stream data. Koken et al. discuss the modelling of ozone depletion and skin cancer rates in Australia. A model is presented to estimate the future increase in skin cancer as a result of an increase in UV radiation. Wang considers two models for the analysis of extreme observations and uses two examples to illustrate their use.
The paper on randomization by Eric Smith was also presented at the 1994 conference in the Symposium on Computer Intensive Methods in Environmental Science organized by Roger Green of The University of Western Ontario and Don A. Jackson of The University of Toronto. The paper discusses the uses of randomization methods in the analysis of multivariate ecological data. Examples from multispecies studies are used for illustration.
Useful economic concepts in setting priorities for environmental policies are presented in the ยฎnal four papers of this issue. The paper by Godby et al. on experimental economic analysis of emissions trading in the presence of market uncertainty considers whether to allow banking and whether to allow trading in entitlement to future permits. The authors conclude that banking and share trading both lead to greater eciency. Baldev Raj and Michael Veall use bootstrapped sensitivity analysis to gain insight into the energyยฑcapital complementarity debate. The paper by Robert Moir recommends the use of an `exact' technique based on randomization in the statistical analysis of economic experimental data. The next paper considers the testing of long-run response in time series environmetrics and econometrics models. Wasif Rasheed and Michael Veall apply their concepts to a simple hydrological model of river ยฏow, a simple electricity demand model, and a number of artiยฎcially constructed data sets designed to illustrate the issues in a Monte Carlo context.