The total cost of misclassification in credit scoring: A comparison of generalized linear models and generalized additive models

2018 ◽  
Vol 38 (5) ◽  
pp. 375-389 ◽  
Author(s):  
Christian Lohmann ◽  
Thorsten Ohliger
Keyword(s):  
Generalized Linear Models ◽  
Linear Models ◽  
Generalized Additive Models ◽  
Credit Scoring ◽  
Additive Models ◽  
Total Cost ◽  
Cost Of Misclassification

Climate Influences on Meningitis Incidence in Northwest Nigeria

2014 ◽  
Vol 6 (1) ◽  
pp. 62-76 ◽  
Author(s):  
Auwal F. Abdussalam ◽  
Andrew J. Monaghan ◽  
Vanja M. Dukić ◽  
Mary H. Hayden ◽  
Thomas M. Hopson ◽  
...  
Keyword(s):  
Generalized Linear Models ◽  
Linear Models ◽  
Generalized Additive Models ◽  
Skill Score ◽  
Additive Models ◽  
Maximum Temperature ◽  
Daily Maximum Temperature ◽  
Daily Maximum ◽  
Explanatory Variables ◽  
Score Statistics

Abstract Northwest Nigeria is a region with a high risk of meningitis. In this study, the influence of climate on monthly meningitis incidence was examined. Monthly counts of clinically diagnosed hospital-reported cases of meningitis were collected from three hospitals in northwest Nigeria for the 22-yr period spanning 1990–2011. Generalized additive models and generalized linear models were fitted to aggregated monthly meningitis counts. Explanatory variables included monthly time series of maximum and minimum temperature, humidity, rainfall, wind speed, sunshine, and dustiness from weather stations nearest to the hospitals, and the number of cases in the previous month. The effects of other unobserved seasonally varying climatic and nonclimatic risk factors that may be related to the disease were collectively accounted for as a flexible monthly varying smooth function of time in the generalized additive models, s(t). Results reveal that the most important explanatory climatic variables are the monthly means of daily maximum temperature, relative humidity, and sunshine with no lag; and dustiness with a 1-month lag. Accounting for s(t) in the generalized additive models explains more of the monthly variability of meningitis compared to those generalized linear models that do not account for the unobserved factors that s(t) represents. The skill score statistics of a model version with all explanatory variables lagged by 1 month suggest the potential to predict meningitis cases in northwest Nigeria up to a month in advance to aid decision makers.

Crash Frequency Analysis with Generalized Additive Models

2008 ◽  
Vol 2061 (1) ◽  
pp. 39-45 ◽  
Author(s):  
Yuanchang Xie ◽  
Yunlong Zhang
Keyword(s):  
Generalized Linear Models ◽  
Linear Models ◽  
Generalized Additive Model ◽  
Negative Binomial ◽  
Generalized Additive Models ◽  
Additive Model ◽  
Additive Models ◽  
Crash Frequency ◽  
Data Set ◽  
Explanatory Variables

Recent crash frequency studies have been based primarily on generalized linear models, in which a linear relationship is usually assumed between the logarithm of expected crash frequency and other explanatory variables. For some explanatory variables, such a linear assumption may be invalid. It is therefore worthwhile to investigate other forms of relationships. This paper introduces generalized additive models to model crash frequency. Generalized additive models use smooth functions of each explanatory variable and are very flexible in modeling nonlinear relationships. On the basis of an intersection crash frequency data set collected in Toronto, Canada, a negative binomial generalized additive model is compared with two negative binomial generalized linear models. The comparison results show that the negative binomial generalized additive model performs best for both the Akaike information criterion and the fitting and predicting performance.

scholarly journals The Evolution of Data Sharing Practices in the Psychological Literature

2021 ◽  
Author(s):  
Judith Neve ◽  
Guillaume A Rousselet
Keyword(s):  
Data Sharing ◽  
Linear Models ◽  
Generalized Additive Models ◽  
Additive Models ◽  
Future Research ◽  
Science Data ◽  
Editorial Policies ◽  
Psychological Science ◽  
The Impact

Sharing data has many benefits. However, data sharing rates remain low, for the most part well below 50%. A variety of interventions encouraging data sharing have been proposed. We focus here on editorial policies. Kidwell et al. (2016) assessed the impact of the introduction of badges in Psychological Science; Hardwicke et al. (2018) assessed the impact of Cognition’s mandatory data sharing policy. Both studies found policies to improve data sharing practices, but only assessed the impact of the policy for up to 25 months after its implementation. We examined the effect of these policies over a longer term by reusing their data and collecting a follow-up sample including articles published up until December 31st, 2019. We fit generalized additive models as these allow for a flexible assessment of the effect of time, in particular to identify non-linear changes in the trend. These models were compared to generalized linear models to examine whether the non-linearity is needed. Descriptive results and the outputs from generalized additive and linear models were coherent with previous findings: following the policies in Cognition and Psychological Science, data sharing statement rates increased immediately and continued to increase beyond the timeframes examined previously, until reaching close to 100%. In Clinical Psychological Science, data sharing statement rates started to increase only two years following the implementation of badges. Reusability rates jumped from close to 0% to around 50% but did not show changes within the pre-policy nor the post-policy timeframes. Journals that did not implement a policy showed no change in data sharing rates or reusability over time. There was variability across journals in the levels of increase, so we suggest future research should examine a larger number of policies to draw conclusions about their efficacy. We also encourage future research to investigate the barriers to data sharing specific to psychology subfields to identify the best interventions to tackle them.

scholarly journals An investigation of human vs. technology-induced variation in catchability for a selection of European fishing fleets

2011 ◽  
Vol 68 (10) ◽  
pp. 2252-2263 ◽  
Author(s):  
Stéphanie Mahévas ◽  
Youen Vermard ◽  
Trevor Hutton ◽  
Ane Iriondo ◽  
Angélique Jadaud ◽  
...  
Keyword(s):  
Linear Models ◽  
Generalized Additive Models ◽  
Technical Information ◽  
Fishing Effort ◽  
Additive Models ◽  
Induced Variation ◽  
Fishing Fleets ◽  
The Impact ◽  
Selection Of ◽  
Experience Effect

Abstract Mahévas, S., Vermard, Y., Hutton, T., Iriondo, A., Jadaud, A., Maravelias, C. D., Punzón, A., Sacchi, J., Tidd, A., Tsitsika, E., Marchal, P., Goascoz, N., Mortreux, S., and Roos, D. 2011. An investigation of human vs. technology-induced variation in catchability for a selection of European fishing fleets. – ICES Journal of Marine Science, 68: 2252–2263. The impact of the fishing effort exerted by a vessel on a population depends on catchability, which depends on population accessibility and fishing power. The work investigated whether the variation in fishing power could be the result of the technical characteristics of a vessel and/or its gear or whether it is a reflection of inter-vessel differences not accounted for by the technical attributes. These inter-vessel differences could be indicative of a skipper/crew experience effect. To improve understanding of the relationships, landings per unit effort (lpue) from logbooks and technical information on vessels and gears (collected during interviews) were used to identify variables that explained variations in fishing power. The analysis was undertaken by applying a combination of generalized additive models and generalized linear models to data from several European fleets. The study highlights the fact that taking into account information that is not routinely collected, e.g. length of headline, weight of otter boards, or type of groundrope, will significantly improve the modelled relationships between lpue and the variables that measure relative fishing power. The magnitude of the skipper/crew experience effect was weaker than the technical effect of the vessel and/or its gear.

scholarly journals Percolation models of pathogen spillover

2019 ◽  
Vol 374 (1782) ◽  
pp. 20180331 ◽  
Author(s):  
Alex D. Washburne ◽  
Daniel E. Crowley ◽  
Daniel J. Becker ◽  
Kezia R. Manlove ◽  
Marissa L. Childs ◽  
...  
Keyword(s):  
Statistical Analysis ◽  
Linear Models ◽  
Mechanistic Model ◽  
Piecewise Linear ◽  
Generalized Additive Models ◽  
Additive Models ◽  
Management Intervention ◽  
Management Concepts ◽  
The Impact ◽  
Reservoir Hosts

Predicting pathogen spillover requires counting spillover events and aligning such counts with process-related covariates for each spillover event. How can we connect our analysis of spillover counts to simple, mechanistic models of pathogens jumping from reservoir hosts to recipient hosts? We illustrate how the pathways to pathogen spillover can be represented as a directed graph connecting reservoir hosts and recipient hosts and the number of spillover events modelled as a percolation of infectious units along that graph. Percolation models of pathogen spillover formalize popular intuition and management concepts for pathogen spillover, such as the inextricably multilevel nature of cross-species transmission, the impact of covariance between processes such as pathogen shedding and human susceptibility on spillover risk, and the assumptions under which the effect of a management intervention targeting one process, such as persistence of vectors, will translate to an equal effect on the overall spillover risk. Percolation models also link statistical analysis of spillover event datasets with a mechanistic model of spillover. Linear models, one might construct for process-specific parameters, such as the log-rate of shedding from one of several alternative reservoirs, yield a nonlinear model of the log-rate of spillover. The resulting nonlinearity is approximately piecewise linear with major impacts on statistical inferences of the importance of process-specific covariates such as vector density. We recommend that statistical analysis of spillover datasets use piecewise linear models, such as generalized additive models, regression clustering or ensembles of linear models, to capture the piecewise linearity expected from percolation models. We discuss the implications of our findings for predictions of spillover risk beyond the range of observed covariates, a major challenge of forecasting spillover risk in the Anthropocene. This article is part of the theme issue ‘Dynamic and integrative approaches to understanding pathogen spillover’.

Sign in / Sign up

Export Citation Format

Share Document