Bias reduction in exponential family nonlinear models

Kosmidis, Ioannis; Firth, David

doi:10.1093/biomet/asp055

Cited by 157 publications

(221 citation statements)

References 17 publications

Supporting

Mentioning

216

Contrasting

Unclassified

Order By: Relevance

“…Regression analyses were performed using the brglm package (Kosmidis 2013), which addresses issues of near perfect separation in logistic regressions (i.e., when there is perfect correspondence of the response variable for most values of the predictors, but not for all) (Heinze and Schemper 2002). All statistical analyses were performed in R (R Development Core Team 2015).…”

Section: Discussionmentioning

confidence: 99%

Adaptation of forest management to climate change as perceived by forest owners and managers in Belgium

et al. 2016

View full text Add to dashboard Cite

Background: Climate change is likely to cause significant modifications in forests. Rising to this challenge may require adaptation of forest management, and therefore should trigger proactive measures by forest managers, but it is unclear to what extent this is already happening. Methods: The survey carried out in this research assesses how forest stakeholders in Belgium perceive the role of their forest management in the context of climate change and the impediments that limit their ability to prepare and respond to these changes. Results: Respondents indicated strong awareness of the changing climate, with more than two-thirds (71 %) expressing concern about the impacts of climate change on their forests. However, less than one-third of the respondents (32 %) reported modifying their management practices motivated by climate change. Among the major constraints limiting their climate related actions, lack of information was considered the most important for managers of both public and private forests. Conclusions: Knowledge transfer is an essential condition for research to lead to innovation. Improving the communication and demonstration of possible solutions for climate change adaptation is therefore likely to be the most effective strategy for increasing their adoption.

show abstract

Section: Discussionmentioning

confidence: 99%

Adaptation of forest management to climate change as perceived by forest owners and managers in Belgium

et al. 2016

View full text Add to dashboard Cite

show abstract

“…This can potentially have a big impact on the reported significance of the explanatory variables. In order to get more realistic estimates of the standard errors of the model parameters, we use the bias reduction method that is supplied in the 'betareg' package (Firth 1993;Kosmidis and Firth 2009).…”

Section: Statistical Model Fitting Techniquementioning

confidence: 99%

Seismic vulnerability functions for Australian buildings by using GEM empirical vulnerability assessment guidelines

et al. 2015

View full text Add to dashboard Cite

Australia has a low to moderate seismicity by world standards. However, the seismic risk is significant due to the legacy of older buildings constructed prior to the national implementation of an earthquake building standard in Australia. The 1989 Newcastle and the 2010 Kalgoorlie earthquakes are the most recent Australian earthquakes to cause significant damage to unreinforced masonry (URM) and light timber frame structures and have provided the best opportunities to examine the earthquake vulnerability of these building types. This paper describes the two above-mentioned building types with a differentiation of older legacy buildings constructed prior to 1945 to the relatively newer ones constructed after 1945. Furthermore, the paper presents method to utilise the large damage and loss-related data (14,000 insurance claims in Newcastle and 400 surveyed buildings in Kalgoorlie) collected from these events to develop empirical vulnerability functions. The method adopted here followed the GEM empirical vulnerability assessment guidelines which involve preparing a loss database, selecting an appropriate intensity measure, selecting and applying a suitable statistical approach to develop vulnerability functions and the identification of optimum functions. The adopted method uses a rigorous statistical approach to quantify uncertainty in vulnerability functions and provides an optimum solution based on goodness-of-fit tests. The analysis shows that the URM structures built before 1945 are the most vulnerable to earthquake with post-1945 URM structures being the next most vulnerable. Timber structures appear to be the least vulnerable, with little difference observed in the vulnerability of timber buildings built before or after 1945. Moreover, the older structures (both URM and timber) exhibit more scatter in results reflecting greater variation in building vulnerability and performance during

show abstract

“…Hence, expression (13) in Kosmidis & Firth (2009) gives that the bias-reducing adjusted score equations using adjustments based on the expected information matrix take the form…”

Section: Bias Reduction Via the Log-linear Modelmentioning

confidence: 99%