A semiparametric negative binomial generalized linear model for modeling over-dispersed count data with a heavy tail: Characteristics and applications to crash data

Shirazi, Mohammadali; Lord, Dominique; Dhavala, Soma S.; Geedipally, Srinivas Reddy

doi:10.1016/j.aap.2016.02.020

Cited by 53 publications

(20 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Meanwhile, considering that over-dispersion is almost always present in crash data and that mixed-effects (fixed and random) could exist, the use of generalized linear mixed models should be considered as a modelling alternative. [19][20][21][22] In Colombia, where the orographic conditions are the main criteria for road design in most cases, there are no studies that evaluate its influence on the crashes. For this reason, we will evaluate several variables and their influence through econometric modelling, including the terrain type as an explanatory variable.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Factors Influencing Crash Frequency on Colombian Rural Roads

Arévalo-Támara

Orozco-Fontalvo

Cantillo

2020

PROMET

View full text Add to dashboard Cite

Traffic crashes in Colombia have become a public health problem causing about 7,000 deaths and 45,000 severe injuries per year. Around 40% of these events occur on rural roads, taking note that the vulnerable users (pedestrians, motorcyclists, cyclists) account for the largest percentage of the victims. The objective of this research is to identify the factors that influence the frequency of crashes, including the singular orography of the country. For this purpose, we estimated Negative Binomial (Poisson-gamma) regression, Zero-inflated model, and generalized the linear mixed model, thus developing a comparative analysis of results in the Colombian context. The data used in the study came from the official sources regarding records about crashes with consequences; that is, with the occurrence of fatalities or injuries on the Colombian roads. For collecting the highway characteristics, an in-field inventory was conducted, gathering information about both infrastructure and operational parameters in more than three thousand kilometres of the national network. The events were geo-referenced, with registries of vehicles, involved victims, and their condition. The results suggest that highways in flat terrain have higher crash frequency than highways in rolling or mountainous terrain. Besides, the presence of pedestrians, the existence of a median and the density of intersections per kilometre also increase the probability of crashes. Meanwhile, roads with shoulders and wide lanes have lower crash frequency. Specific interventions in the infrastructure and control for reducing crashes risk attending the modelling results have been suggested.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Mientras tanto, las carreteras con berma y calles anchas tienen menos frecuencia de choques. Como recomendación, se sugieren intervenciones result is consistent with the findings of Shirazi et al [19]. In this case, the use of a more complex model (GLMM) did not provide a better fit.…”

Section: Factores Influyentes En La Frecuencia De Choques En Las Víasunclassified

Factors Influencing Crash Frequency on Colombian Rural Roads

Arévalo-Támara

Orozco-Fontalvo

Cantillo

2020

PROMET

View full text Add to dashboard Cite

show abstract

“…Using boosting with GLMs can improve prediction accuracy 23 . We applied two GLMs; boosting GLM (BGLM) and negative binomial GLM (NBGLM) 51 .…”

Section: Generalized Linear Models (Glms)mentioning

confidence: 99%

Mapping wind erosion hazard with regression-based machine learning algorithms

Gholami

Mohammadifar

Bui

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Land susceptibility to wind erosion hazard in Isfahan province, Iran, was mapped by testing 16 advanced regression-based machine learning methods: Robust linear regression (RLR), Cforest, Non-convex penalized quantile regression (NCPQR), Neural network with feature extraction (NNFE), Monotone multi-layer perception neural network (MMLPNN), Ridge regression (RR), Boosting generalized linear model (BGLM), Negative binomial generalized linear model (NBGLM), Boosting generalized additive model (BGAM), Spline generalized additive model (SGAM), Spike and slab regression (SSR), Stochastic gradient boosting (SGB), support vector machine (SVM), Relevance vector machine (RVM) and the Cubist and Adaptive network-based fuzzy inference system (ANFIS). Thirteen factors controlling wind erosion were mapped, and multicollinearity among these factors was quantified using the tolerance coefficient (TC) and variance inflation factor (VIF). Model performance was assessed by RMSE, MAE, MBE, and a Taylor diagram using both training and validation datasets. The result showed that five models (MMLPNN, SGAM, Cforest, BGAM and SGB) are capable of delivering a high prediction accuracy for land susceptibility to wind erosion hazard. DEM, precipitation, and vegetation (NDVI) are the most critical factors controlling wind erosion in the study area. Overall, regression-based machine learning models are efficient techniques for mapping land susceptibility to wind erosion hazards.

show abstract

“…The over-dispersion characteristic in discrete count data can be captured by Negative Binomial (NB) or Poisson-Gamma (PG) mixture density. A flexible NB generalized linear model for over-dispersed count data was proposed by Shirazi et al [11] with randomly distributed mixed effects characterized by either Lindley distribution or Dirichlet Process (DP). Si et al [12] derive a Poisson and Negative Binomial model-based clustering algorithms for RNA-seq count data to group genes with similar expression level per treatment, using Expectation-Maximization (EM) algorithm along with initialization technique and stochastic annealing algorithm.…”

Section: Introductionmentioning

confidence: 99%

Subject level clustering using a negative binomial model for small transcriptomic studies

Noel-Macdonnell

Koestler

et al. 2018

BMC Bioinformatics

View full text Add to dashboard Cite

BackgroundUnsupervised clustering represents one of the most widely applied methods in analysis of high-throughput ‘omics data. A variety of unsupervised model-based or parametric clustering methods and non-parametric clustering methods have been proposed for RNA-seq count data, most of which perform well for large samples, e.g. N ≥ 500. A common issue when analyzing limited samples of RNA-seq count data is that the data follows an over-dispersed distribution, and thus a Negative Binomial likelihood model is often used. Thus, we have developed a Negative Binomial model-based (NBMB) clustering approach for application to RNA-seq studies.ResultsWe have developed a Negative Binomial Model-Based (NBMB) method to cluster samples using a stochastic version of the expectation-maximization algorithm. A simulation study involving various scenarios was completed to compare the performance of NBMB to Gaussian model-based or Gaussian mixture modeling (GMM). NBMB was also applied for the clustering of two RNA-seq studies; type 2 diabetes study (N = 96) and TCGA study of ovarian cancer (N = 295). Simulation results showed that NBMB outperforms GMM applied with different transformations in majority of scenarios with limited sample size. Additionally, we found that NBMB outperformed GMM for small clusters distance regardless of sample size. Increasing total number of genes with fixed proportion of differentially expressed genes does not change the outperformance of NBMB, but improves the overall performance of GMM. Analysis of type 2 diabetes and ovarian cancer tumor data with NBMB found good agreement with the reported disease subtypes and the gene expression patterns. This method is available in an R package on CRAN named NB.MClust.ConclusionUse of Negative Binomial model based clustering is advisable when clustering over dispersed RNA-seq count data.Electronic supplementary materialThe online version of this article (10.1186/s12859-018-2556-9) contains supplementary material, which is available to authorized users.

show abstract

A semiparametric negative binomial generalized linear model for modeling over-dispersed count data with a heavy tail: Characteristics and applications to crash data

Cited by 53 publications

References 36 publications

Factors Influencing Crash Frequency on Colombian Rural Roads

Factors Influencing Crash Frequency on Colombian Rural Roads

Mapping wind erosion hazard with regression-based machine learning algorithms

Subject level clustering using a negative binomial model for small transcriptomic studies

Contact Info

Product

Resources

About