Zero-inflated Count Research Articles

Patent citations and knowledge spillovers: an analysis of Chinese patents registered in the USA

This paper examines US patent citation data and analyses how different firms in China affect knowledge spillovers. Patents granted by the US patent office to applicants located in China are collected along with their citation counts. Two kinds of patent citations, namely citations of previous patents and those of non-patent literature, are used to measure knowledge flows. In the empirical analysis, the negative binomial and zero-inflated count models are considered. The regression results suggest the existence of heterogeneity among firms of different ownership. In terms of knowledge spillovers, the US multinational corporations perform better than those from other western countries; Taiwanese companies outperform their counterparts from Hong Kong and Chinese private corporations contribute more than Chinese state-owned enterprises. These results have important policy implications for the development of a knowledge-intensive economy in China.

Zero adjusted models with applications to analysing helminths count data.

BackgroundIt is common in public health and epidemiology that the outcome of interest is counts of events occurrence. Analysing these data using classical linear models is mostly inappropriate, even after transformation of outcome variables due to overdispersion. Zero-adjusted mixture count models such as zero-inflated and hurdle count models are applied to count data when over-dispersion and excess zeros exist. Main objective of the current paper is to apply such models to analyse risk factors associated with human helminths (S. haematobium) particularly in a case where there’s a high proportion of zero counts.MethodsThe data were collected during a community-based randomised control trial assessing the impact of mass drug administration (MDA) with praziquantel in Malawi, and a school-based cross sectional epidemiology survey in Zambia. Count data models including traditional (Poisson and negative binomial) models, zero modified models (zero inflated Poisson and zero inflated negative binomial) and hurdle models (Poisson logit hurdle and negative binomial logit hurdle) were fitted and compared.ResultsUsing Akaike information criteria (AIC), the negative binomial logit hurdle (NBLH) and zero inflated negative binomial (ZINB) showed best performance in both datasets. With regards to zero count capturing, these models performed better than other models.ConclusionThis paper showed that zero modified NBLH and ZINB models are more appropriate methods for the analysis of data with excess zeros. The choice between the hurdle and zero-inflated models should be based on the aim and endpoints of the study.

BAYESIAN SPATIAL-TEMPORAL MODELING OF ECOLOGICAL ZERO-INFLATED COUNT DATA.

A Bayesian hierarchical model is developed for count data with spatial and temporal correlations as well as excessive zeros, uneven sampling intensities, and inference on missing spots. Our contribution is to develop a model on zero-inflated count data that provides flexibility in modeling spatial patterns in a dynamic manner and also improves the computational efficiency via dimension reduction. The proposed methodology is of particular importance for studying species presence and abundance in the field of ecological sciences. The proposed model is employed in the analysis of the survey data by the Northeast Fisheries Sciences Center (NEFSC) for estimation and prediction of the Atlantic cod in the Gulf of Maine - Georges Bank region. Model comparisons based on the deviance information criterion and the log predictive score show the improvement by the proposed spatial-temporal model.

Group Comparisons Involving Zero-Inflated Count Data in Clinical Trials

Group Comparisons Involving Zero-Inflated Count Data in Clinical Trials

Variable selection approach for zero-inflated count data via adaptive lasso

This article proposes a variable selection approach for zero-inflated count data analysis based on the adaptive lasso technique. Two models including the zero-inflated Poisson and the zero-inflated negative binomial are investigated. An efficient algorithm is used to minimize the penalized log-likelihood function in an approximate manner. Both the generalized cross-validation and Bayesian information criterion procedures are employed to determine the optimal tuning parameter, and a consistent sandwich formula of standard errors for nonzero estimates is given based on local quadratic approximation. We evaluate the performance of the proposed adaptive lasso approach through extensive simulation studies, and apply it to analyze real-life data about doctor visits.

On Baseline Conditions for Zero-Inflated Longitudinal Count Data

We describe a mixed-effect hurdle model for zero-inflated longitudinal count data, where a baseline variable is included in the model specification. Association between the count data process and the endogenous baseline variable is modeled through a latent structure, assumed to be dependent across equations. We show how model parameters can be estimated in a finite mixture context, allowing for overdispersion, multivariate association and endogeneity of the baseline variable. The model behavior is investigated through a large-scale simulation experiment. An empirical example on health care utilization data is provided.

Analyzing propensity matched zero-inflated count outcomes in observational studies

Determining the effectiveness of different treatments from observational data, which are characterized by imbalance between groups due to lack of randomization, is challenging. Propensity matching is often used to rectify imbalances among prognostic variables. However, there are no guidelines on how appropriately to analyze group matched data when the outcome is a zero-inflated count. In addition, there is debate over whether to account for correlation of responses induced by matching and/or whether to adjust for variables used in generating the propensity score in the final analysis. The aim of this research is to compare covariate unadjusted and adjusted zero-inflated Poisson models that do and do not account for the correlation. A simulation study is conducted, demonstrating that it is necessary to adjust for potential residual confounding, but that accounting for correlation is less important. The methods are applied to a biomedical research data set.

Modeling zero‐inflated count data when exposure varies: With an application to tumor counts

This paper is concerned with the analysis of zero-inflated count data when time of exposure varies. It proposes a modified zero-inflated count data model where the probability of an extra zero is derived from an underlying duration model with Weibull hazard rate. The new model is compared to the standard Poisson model with logit zero inflation in an application to the effect of treatment with thiotepa on the number of new bladder tumors.

Herd-level risk factors for Campylobacter fetus infection, Brucella seropositivity and within-herd seroprevalence of brucellosis in cattle in northern Nigeria

Brucellosis and campylobacteriosis are economically important diseases affecting bovine reproductive efficiency in Nigeria. A questionnaire-based survey was conducted in 271 cattle herds in Adamawa, Kaduna and Kano states of northern Nigeria using multistage cluster sampling. Serum from 4745 mature animals was tested for Brucella antibodies using the Rose-Bengal plate test and positives were confirmed in series-testing protocol using competitive enzyme-linked immunosorbent assay. Preputial scrapings from 602 bulls were tested using culture and identification for Campylobacter fetus. For each disease, a herd was classified as positive if one or more animals tested positive. For each herd, information on potential managemental and environmental risk factors was collected through a questionnaire administered during an interview with the manager, owner or herdsman. Multiple logistic regression models were used to model the odds of herd infection for each disease. A zero-inflated Poisson model was used to model the count of Brucella-positive animals within herds, with the number tested as an exposure variable. The presence of small ruminants (sheep and/or goats) on the same farm, and buying-in of >3 new animals in the previous year or failure to practice quarantine were associated with increased odds of herd-level campylobacteriosis and brucellosis, as well as increased within-herd counts of Brucella-positive animals. In addition, high rainfall, initial acquisition of animals from markets, practice of gynaecological examination and failure to practice herd prophylactic measures were positively associated with the odds of C. fetus infection in the herd. Herd size of >15, pastoral management system and presence of handling facility on the farm were associated with increased odds, and gynaecological examination with reduced odds of herd-level Brucella seropositivity. Furthermore, the zero-inflated Poisson model showed that borrowing or sharing of bulls was associated with higher counts, and provision of mineral supplement with lower counts of Brucella-positive cattle within herds. Identification of risk factors for bovine campylobacteriosis and brucellosis can help to identify appropriate control measures, and the use of zero-inflated count model can provide more specific information on these risk factors.

A dynamic hurdle model for zeroinflated panel count data

This article proposes an approximate conditional dynamic finite mixture hurdle model for panel count data with excess of zeros and endogenous initial conditions. We provide parameter estimates by using the Expectation-Maximization (EM) algorithm in a Nonparametric Maximum Likelihood (NPML) framework. An application to a unique data set on traffic violation counts of a subpopulation of Italian drivers is given.

How Managed Care Affects Medicaid Utilization: A Synthetic Difference-in-Differences Zero-Inflated Count Model

We develop a synthetic difference-in-differences statistical design to apply to experimental data for adult women living in Hennepin County, Minnesota, to estimate the impact of Medicaid managed care on various modes of medical care use.Because the outcomes of interest are utilization counts with many persons using none of a particular mode of care we use count regression models that are adjusted for excessive zeros.We find no reductions in physician visits or hospital inpatient and emergency department care use, but reductions in hospital outpatient care.Simulations designed to judge the economic significance of our results suggest a program effect that is a savings of about ten percent.

Semiparametric Analysis of Longitudinal Zero-inflated Count Dat a with Applications to Instrumental Activities of Daily Living

Background: The instrumental activities of daily living (IADLs) are important index of physical functioning in older adult studies. These count outcomes with a large proportion of zeros are often collected in longitudinal studies. Data were from the Hispanic Established Population for Epidemiological Study of the Elderly (HEPESE), a four wave (seven years) longitudinal study of community-dwelling elderly Mexican-Americans. There were excess zeros IADLs observed during follow-up. Methods: We present semiparametric zero-inflated Poisson (ZIP) and hurdle model with random effects to evaluate IADLs in the context of excess zeros. Results: Age, education, household income, marital status, smoking, cognitive functioning and prescription medication use were not significantly associated with IADLs. The counts of IADLs changed with age nonlinearly. Conclusions: zero-inflated Poisson (ZIP) and hurdle model with random effect fits the IADLs counts with excess zeros in longitudinal studies.

Social Learning through Social Networks and Technological Appropriation: The Role of Peer Effects in the Adoption and Use of the Internet in Cameroonian Tontines

This article attempts to verify whether tontines constitute a social learning platform for the appropriation of technology that underlies patterns of adoption and use of the Internet in sub-Saharan Africa. It uses for this purpose a survey of 2650 households in the cities of Douala, Buea and Limbe and employs initially two zero-inflated count data models to highlight the Intensity of the practice of tontines and after, simple probit models to detect the presence of favorable peer effects in the association tontines on the likelihood of adoption and use of the Internet in Cameroonian households. The results show that, while in ROSCA-type tontines social learning is uncertain because of neutral peer effects, in association tontines within socio-professional groups peer effects are positive and reinforce the process of adoption. These results confirm the studies on dissocializing Internet carried in United States and Europe at the beginning of the dynamics of its diffusion. However and more important, the results presented here show that peer effects play positively on social learning and technology appropriation that underlie the adoption and use of the Internet not in tontine of ROSCA-type, but in the tontine envisaged in professional groups according to their propensity to use modern values for jobs.

Sinners or Saints? Preachers' Kids and Risky Health Behaviors

This paper examines parental influence on adolescent risky behavior, focusing on a unique population: children of the clergy, more commonly known as preachers' kids (PKs). We use latent variable and zero-inflated count models to analyze the effect of being a PK on both uptake and intensity of use of alcohol, cigarettes, marijuana, and other drugs. We find that being a PK significantly reduces alcohol use. This effect comes exclusively from a reduction in the probability of any alcohol use and this increased abstinence among children of the clergy persists into adulthood. These results are consistent with popular conceptions that PKs either take no risks or take large risks. We find no significant effects of being a PK on cigarette uptake or intensity of use but some evidence of a negative PK effect on the uptake of marijuana and other drugs.

What belongs where? Variable selection for zero-inflated count models with an application to the demand for health care

This paper develops a Bayesian spike and slab model for zero-inflated count models which are commonly used in health economics. We account for model uncertainty and allow for model averaging in situations with many potential regressors. The proposed techniques are applied to a German data set analyzing the demand for health care. An accompanying package for the free statistical software environment R is provided.

Statistical models for longitudinal zero‐inflated count data with applications to the substance abuse field

This study fills in the current knowledge gaps in statistical analysis of longitudinal zero-inflated count data by providing a comprehensive review and comparison of the hurdle and zero-inflated Poisson models in terms of the conceptual framework, computational advantage, and performance under different real data situations. The design of simulations represents the special features of a well-known longitudinal study of alcoholism so that the results can be generalizable to the substance abuse field. When the hurdle model is more natural under the conceptual framework of the data, the zero-inflated Poisson model tends to produce inaccurate estimates. Model performance improves with larger sample sizes, lower proportions of missing data, and lower correlations between covariates. The simulation also shows that the computational strength of the hurdle model disappears when random effects are included.

Analysis of Zero-Inflated Count Data From Clinical Trials With Potential Dropouts

Counts of prespecified events are important endpoints for many safety and efficacy clinical trials. The conventional Poisson model might not be ideal due to three potential issues: (1) overdispersion arising from intra-subject correlation, (2) zero inflation when the prespecified event is rare, and (3) missing observations due to early dropouts. Negative binomial (NB), Poisson hurdle (PH), and negative binomial hurdle (NBH) models are more appropriate for overdispersed and/or zero-inflated count data. An offset can be included in these models to adjust for differential exposure duration due to early dropouts. In this article, we propose new link functions for the hurdle part of a PH/NBH model to facilitate testing for zero-inflation and model selection. The proposed link function particularly improves the model fit of a NBH model when an offset is included to adjust for differential exposure. A simulation study is conducted to compare the existing and proposed models, which are then applied to data from two clinical trials to demonstrate application and interpretation of these methods.

Score Tests for Semiparametric Zero-inflated Poisson Models

Count data sets often produce many zeros. It is sometimes potentially questionable to use a linear predictor to model the effect of a continuous covariate of interest in zero-inflated count data. To relax the restriction, Li (2011) proposed a semiparametric zero-inflated Poisson (ZIP) regression model by using fixed-knot cubic $basis$ splines or $B$-splines to model the covariate effect, and used the likelihood ratio test to assess the validity of the linear relationship between the natural logarithm of the Poisson mean and the covariate. A score test is conducted to assess whether the extra proportion of zeros in the semiparametric ZIP regression model is equal to zero.

CONSISTENT ESTIMATION OF ZERO‐INFLATED COUNT MODELS

Applications of zero-inflated count data models have proliferated in health economics. However, zero-inflated Poisson or zero-inflated negative binomial maximum likelihood estimators are not robust to misspecification. This article proposes Poisson quasi-likelihood estimators as an alternative. These estimators are consistent in the presence of excess zeros without having to specify the full distribution. The advantages of the Poisson quasi-likelihood approach are illustrated in a series of Monte Carlo simulations and in an application to the demand for health services.

The k‐ZIG: Flexible Modeling for Zero‐Inflated Counts

Many applications involve count data from a process that yields an excess number of zeros. Zero-inflated count models, in particular, zero-inflated Poisson (ZIP) and zero-inflated negative binomial (ZINB) models, along with Poisson hurdle models, are commonly used to address this problem. However, these models struggle to explain extreme incidence of zeros (say more than 80%), especially to find important covariates. In fact, the ZIP may struggle even when the proportion is not extreme. To redress this problem we propose the class of k-ZIG models. These models allow more flexible modeling of both the zero-inflation and the nonzero counts, allowing interplay between these two components. We develop the properties of this new class of models, including reparameterization to a natural link function. The models are straightforwardly fitted within a Bayesian framework. The methodology is illustrated with simulated data examples as well as a forest seedling dataset obtained from the USDA Forest Service's Forest Inventory and Analysis program.