Bayesian regression filter and the issue of priors

We propose a Bayesian framework for regression problems, which covers areas which are usually dealt with by function approximation. An online learning algorithm is derived which solves regression problems with a Kalman filter. Its solution always improves with increasing model complexity, without the risk of over-fitting. In the infinite dimension limit it approaches the true Bayesian posterior. The issues of prior selection and over-fitting are also discussed, showing that some of the commonly held beliefs are misleading. The practical implementation is summarised. Simulations using 13 popular publicly available data sets are used to demonstrate the method and highlight important issues concerning the choice of priors.

Special issue of PDS - Based on PDDG meeting, Aston University, September 2007, in honour of Professor Norman Billingham

This edition of Polymer Degradation and Stability comprises selected papers presented at the 27th Polymer Degradation Discussion Group (PDDG) Conference, affiliated to the Macromolecular Group of the Royal Society of Chemistry, held at Aston University, Birmingham (UK), September 2007, in honour of Professor Norman Billingham for lifelong career in Polymer Science. Norman Billingham was one of the main founders of the PDDG meetings which have been held annually at different venues in the UK since 1976, and biennially since 1998. The PDDG meetings are forums for scientific developments and open discussion of all aspects of polymer degradation and stabilisation and have always provided younger researchers a platform for presenting and discussing their work in the field. This 27th PDDG meeting, chaired by Sahar Al-Malaika (Aston University), was dedicated to Norman Billingham, the Guest of Honour, in recognition for his distinguished contributions to Polymer Science particularly in the area of polymer degradation and stabilisation. The meeting was attended by 65 delegates from 15 countries in Europe, USA and Australia, many of whom have worked and collaborated with Norman at some points of their career. The guest editors of this issue wish to add their Congratulations to Norman Billingham for his outstanding contributions and achievements in the field of polymer degradation and stabilisation and to wish him very happy years ahead. We wish also to express our gratitude to Elsevier for sponsoring the 27th PDDG event and our appreciation to Professor Norman Billingham in his capacity as the Editor-in-Chief of Polymer Degradation and Stability, for the support and help received during the preparation of this special issue.

Special issue of accounting forum:social and environmental accounting and reporting in emerging and less developed countries

Call for papers

Editorial:welcome to the LRN2004 annual conference special issue of the International journal of logistics: research and applications

It is a great pleasure to be Guest Editor for this issue – I hope that the papers which are included will be stimulating and support you in your ongoing research activities. A number of guiding principles were adopted in selecting the papers for inclusion in this issue. Firstly, the papers cover a wide range of logistics and supply chain management (SCM) topics. This is a reflection of the evolution of the field in recent years. In terms of the “buy-make-store-move-sell” model of SCM all the main constituent areas are addressed. Secondly, it is important that the conference issue of this Journal reflects the emphasis and content of the conference itself. I have tried to achieve this in terms of the papers included. One interesting point to note is that outsourcing is a theme which is a major issue in a number of papers. This reflects the increasing importance of this issue to organisations of all kinds and sizes. Economic globalisation and the trend towards vertical disintegration of supply chain architectures have sharpened the focus on outsourcing as a key element of supply chain strategy. The need to move beyond the notion that sourcing of certain activities can be some kind of panacea in evident from the relevant contributions. Thirdly, the LRN Annual Conference has become a more international event in recent years...the number of delegates and papers presented from outside the UK has continued to grow. The papers collected in this issue reflect this internationalization. Two papers are worthy of particular comment from an LRN perspective. The contribution by Jaafar and Rafiq has been developed from the submission which won the best paper prize at the LRN 2004 event. The paper by Pettit and Beresford is based on research which was supported by LRN seed corn funding. It was developed form the final report on this work submitted to CITL (UK) via the LRN. The seed corn funding is an important mechanism whereby the LRN supports research in innovative aspects of logistics in UK universities. In many ways, the LRN2004 event in Dublin seems like a long time ago. From my point of view it was one of the most professionally rewarding activities in which I have been involved in my career. It was a time to meet old friends and new and to keep abreast of the multitude of interesting projects being undertaken in over 20 countries. There are too many people to thank for the smooth running of the event. However, my colleague John Mee does warrant a special mention. His logistical skills were seriously put to the test in the weeks and months leading up to September 9th. 2004. I want to acknowledge his particular contribution to the success of the event. Since then we have had the 2005 event at the University of Plymouth. This was again a great opportunity to network with colleagues and many congratulations are due to John Dinwoodie and his team. We now look forward to LRN 2006 in Newcastle...form my part I hope and trust that this issue provides some useful perspectives and insights into the range of topics addressed.

Eco-efficiency considering the issue of heterogeneity among power plants

One of the main objectives in restructuring power industry is enhancing the efficiency of power facilities. However, power generation industry, which plays a key role in the power industry, has a noticeable share in emission amongst all other emission-generating sectors. In this study, we have developed some new Data Envelopment Analysis models to find efficient power plants based on less fuel consumption, combusting less polluting fuel types, and incorporating emission factors in order to measure the ecological efficiency trend. We then applied these models to measuring eco-efficiency during an eight-year period of power industry restructuring in Iran. Results reveal that there has been a significant improvement in eco-efficiency, cost efficiency and allocative efficiency of the power plants during the restructuring period. It is also shown that despite the hydro power plants look eco-efficient; the combined cycle ones have been more allocative efficient than the other power generation technologies used in Iran.

Evaluation of public housing strategies in Mexico City : an exploratory goal programming approach

This thesis describes the development and use of a goal programming methodology for the evaluation of public housjng strategies in Mexico City, The methodology responds to the need to incorporate the location, size and densities of housing projects on the one hand, and "external" constraints such as the ability of low income families to pay for housing, and the amounts of capital and land available, on the other. The provision of low cost housing by public housing agencies in Mexico City is becoming increasingly difficult because there are so many constraints to be met and overcome, the most important of which is the ability of families to pay for housing. Other important limiting factors are the availability of capital and of land plots of the right size in desired locations. The location of public housing projects is significant because it determines the cost and pattern of work trips, which in a metropolitan area such as Mexico City are of considerable importance to both planners and potential. house owners. In addition, since the price of land is closely related to its location, the last factor is also significant in determining the price of the total housing package. Consequently there is a major trade-off between a housing strategy based on the provision of housing at locations close to employment, and the opposite one based on the provjsion of housjng at locations where employment accessibility is poorer but housing can be provided at a lower price. The goal programming evaluation methodology presented in this thesis was developed to aid housing planners to evaluate housing strategies which incorporate the issues raised above,