count data models
Recently Published Documents

Background: The objective of this study is comparing different count data models for stillbirth data. In modeling this type of data, Poisson regression or alternative models can be preferred. Methods: The poisson, negative binomial, zero-inflated poisson, zero-inflated negative binomial, poisson-logit hurdle and negative binomial-logit hurdle regressions were compared and used to examine the effects of the gender, parity and herd-year-season independent variables on stillbirth. Furthermore, the Log-Likelihood statistics, Akaike Information Criteria, Bayesian Information Criteria and rootogram graphs were used as comparison criteria for performance of the models. According to these criteria, Negative Binomial-Logit Hurdle Regression model was chosen as the best model. Result: The parameter estimates obtained by Negative Binomial-Logit Hurdle Regression model in relation to the effects of the gender, parity and herd-year-season independent variables on stillbirth were found to be significant (p less than 0.01). It was found that while stillbirth incidence was higher in males than females, it was found to decrease as the parity increased. As a result, the Negative Binomial Logit Hurdle model was found the best model for stillbirth count data with overdispersion.

A Flash Crowd (FC) event occurs when network traffic increases suddenly due to a specific reason (e.g. e-commerce sale). Despite its legitimacy, this kind of situation usually decreases the network resource performance. Furthermore, attackers may simulate FC situations to introduce undetected attacks, such as Distributed Denial of Service (DDoS), since it is very difficult to distinguish between legitimate and malicious data flows. To differentiate malicious and legitimate traffic we propose applying zero inflated count data models in conjunction with the Correlation Coefficient Flow (CCF) method – a well-known method used in FC situations. Our results were satisfactory and improve the accuracy of CCF method. Furthermore, since the environment toggles between normal and FC situations, our method has the advantage of working in both situations.

There is a vast amount of geo-referenced data in many fields of study including ecological studies. Geo-referencing is usually by point referencing; that is, latitudes and longitudes or by areal referencing, which includes districts, counties, states, provinces and other administrative units. The availability of large geo-referenced datasets for modelling has necessitated the development and application of spatial statistical methods. However, spatial varying coefficients models exploring the abundance of tick counts remain limited. In this study we used data that was collected and prepared by researchers in the Department of Biological Sciences from the Old Dominion University, Virginia, USA. We modelled tick life-stage counts and abundance variability from 12 sampling locations, with 5 different habitats (numbered 1-5), three habitat types; namely: woods, edges and grass; collected monthly from May 2009 through December 2018. Spatio-temporal Poisson and spatio-temporal negative binomial (NB) count data models were fitted to the data and compared using the deviance information criteria (DIC). The NB model outperformed the Poisson models with all its DIC values being smaller than those of the Poisson model. Results showed that the covariates varied spatially across counties. There was a decreasing time (in years) effect over the study period. However, even though the time effect was decreasing over the study period, space-time interaction effects were seen to be increasing over time in York County.

This paper examines the determinants of female board representation for a sample of Brazilian listed companies for the year of 2018. Using count data models, we find that greater firm size, performance and board size lead to higher woman representation on companies’ boards. Also, that private control is associated with a lower number of women on boards. Most studies related to board composition focus on independent directors and are conducted in a developed countries’ setting. This work contributes to the extant literature in understanding what drives woman representation on corporate boards in an emerging market context and also would help to support the definition and implementation of gender diversity policies by showing possible impacts.

Using data on greenfield investment in German districts from 2003 to 2010, we examine how regional development policies affect the decision of multinational corporations to locate facilities in Germany. We are interested in whether regional development policies accumulate to increase the attractiveness of a region and whether some policies are necessary to attract foreign investors. Applying count data models and geographic weighted regression, the results indicate that, on average, regional development policies increase the attractiveness of German districts for multinational firms. We find that place-based policies have the strongest effect on investments in the East German lagging regions. However, policies predominantly attract standardised types of investments that require considerable capital investments but not specialised location advantages.

Each year more than three thousand people die and get serious injuries in traffic accidents. Count data model provide more precise tools for planners and decision makers to conduct proactive road safety planning.We analyzed the exploratory research of Road Traffic Accidents (RTAs) and furthermore explores the factors affecting the RTAs frequency in 36 districts of the Punjab over a time period of three years (July 1, 2013 June 30, 2016) with monthly data using panel count data models. Among the models considered, the random parameters Poisson panel count data model is found to fit the data best. The exploratory analysis shows that highly dense populated districts with large number of registered vehicles causes more accidents as compared to low density populated districts. It is found that, most of the variables used to control the variation in the frequency of RTAs counts play vital role with higher significance levels. The application of regression analysis and modeling of RTAs at district level in Punjab will help to identification of districts with high RTAs rates and this could help more efficient road safety management in the Punjab.

PurposeThe World Health Organization issued its global action plan on physical activities 2018–2030, emphasizing the importance of context-specific evidence on the subject. Accordingly, this study aims to provide unique and important policy insights on trends and drivers of participation in physical exercises by academic community in Sri Lankan universities.Design/methodology/approachFor this purpose, we collected cross-sectional data (n = 456) in 2020 using a survey, and first, estimated a double-hurdle model to uncover covariates influencing likelihood and intensity of physical exercises overall. Second, count-data models are estimated to capture regularity of key exercises.FindingsThe results reveal that about 50% of members do not participate in any general physical exercise. Older members (marginal effect (ME) = 3.764, p < 0.01), non-Buddhists (ME = 54.889, p < 0.01) and alcohol consumers (ME = 32.178, p < 0.05) exhibit a higher intensity of participating in exercises overall. The intensity is lower for rural members (ME = −63.807, p < 0.01) and those with health insurance covers (ME = −31.447, p < 0.05). Individuals diagnosed for chronic illnesses show a higher likelihood of exercising but, their time devotion is limited. The number of children the academic staff members have as parents reduces the likelihood, but for those who choose to exercise have higher time devotion with increased number of children. The covariates play a similar role in determining regularity of key exercises: walking, jogging and exercising on workout machines.Research limitations/implicationsThe results imply a need to promote exercising in general and particularly among younger, healthy, insured and female individuals living in rural sector.Originality/valueThe study covers an under-researched professional sub-group in an under-researched developing context, examining both the likelihood and regularity of exercising as both dimensions are equally important for individuals to maintain healthy lives.

Count data appears in many research fields and exhibits certain features that make modeling difficult. Most popular approaches to modeling count data can be classified into observation and parameter-driven models. In this paper, we review two models from these classes: the log-linear multivariate conditional intensity model (also referred to as an integer-valued generalized autoregressive conditional heteroskedastic model) and the non-linear state-space model for count data. We compare these models in terms of forecasting performance on simulated data and two real datasets. In simulations, we consider the case of model misspecification. We find that both models have advantages in different situations, and we discuss the pros and cons of inference for both models in detail.

Abstract For the modelling of count data, aggregation of the raw data over certain subgroups or predictor configurations is common practice. This is, for instance, the case for count data biomarkers of radiation exposure. Under the Poisson law, count data can be aggregated without loss of information on the Poisson parameter, which remains true if the Poisson assumption is relaxed towards quasi-Poisson. However, in biodosimetry in particular, but also beyond, the question of how the dispersion estimates for quasi-Poisson models behave under data aggregation have received little attention. Indeed, for real data sets featuring unexplained heterogeneities, dispersion estimates can increase strongly after aggregation, an effect which we will demonstrate and quantify explicitly for some scenarios. The increase in dispersion estimates implies an inflation of the parameter standard errors, which, however, by comparison with random effect models, can be shown to serve a corrective purpose. The phenomena are illustrated by γ-H2AX foci data as used for instance in radiation biodosimetry for the calibration of dose-response curves.