Zero-inflated Ordered Probit Model Research Articles

Human rights organizations, equal opportunity, and violence against refugees

What factors influence levels of repression targeting refugee populations? In this paper, we explore the efficacy of top-down and bottom-up approaches to mitigating host-state repression, specifically highlighting the role of human rights organizations (HROs) and institutionalized equal opportunity in influencing states’ decision to violently repress refugees. In so doing, we argue that repression levels can be moderated through proper accommodation and integration of refugee populations via the activities of HROs in addition to increased institutionalized equal opportunities for displaced persons. Using a new dataset that captures state repression of refugee populations together with a zero-inflated ordered probit model and an estimation technique that addresses endogeneity, we find mixed support for our hypotheses.

Top-Category Inflation in Ordered International Relations Outcomes

Abstract Ordered dependent variables are widely employed in international relations (IR). These ordered dependent variables often suffer from inflated observations in their highest outcome category due to distinct processes. The application of standard ordered probit and ordered logit models to such ordinal measures will fail to capture these distinct processes, often producing biased inferences as a result. Yet IR researchers have thus far ignored the potential for top-category inflation in ordered outcome variables. We sensitize researchers to top-category inflation in ordered IR outcomes. We then intuitively extend the widely used zero-inflated ordered probit model to the top-category inflated setting, providing resources to facilitate the proper modeling of top-category inflation in ordered outcomes. Finally, we provide two applications to published IR research related to trade politics and political repression. Together, these applications illustrate the substantive and methodological potentials of our proposed tools for diagnosing and modeling top-category inflation in IR outcomes.

Analyzing the Relationships Between Frequency of Leisure Activity Participation and Social Capital

Transportation research has focused on gaining a better understanding of how the need to travel is influenced by one’s social capital and relates to participation in various leisure activities. This study postulated that increased participation frequency of social-oriented leisure activities is an expressive outcome and thus mostly affected by expressive resources of social capital. To explore the impacts of social capital on leisure activity behavior—particularly the participation frequency—a refined survey was designed to obtain an extensive collection of leisure activities, social capital measures, as well as mobility and sociodemographic characteristics. The survey included questions to ascertain individuals’ instrumental and expressive social capital through position, resource, and generalized name generators. The 20 activities that were most reported by the survey respondents were examined using a zero-inflated ordered probit model with correlated error terms. Respondents with greater expressive resources participated in social leisure activities more frequently than those with fewer expressive resources. However, this relationship did not hold for the activities of drinking and socializing, attending church, or dining out. In contrast, instrumental social capital had mixed effects (i.e., positive, negative, or insignificant) in relation to the participation frequency of social activities.

The determinants of commute mode usage frequency of post-secondary students in the Greater Toronto and Hamilton Area

An important aspect of post-secondary student travel behaviour is their commute mode usage frequencies. How frequently students use different transportation modes for their daily travel directly characterizes their habits, routines, and predispositions, which can ultimately affect long-term social welfare of the region, congestion of the transportation network, and fuel consumption. Thus, obtaining an in-depth and unbiased understanding of the various factors influencing this travel behavior is key to the sustainable transportation development of a region. Existing studies on this topic are not comprehensive enough in terms of the types of commute modes analyzed and most of them relied on relatively small samples for their investigation. This paper attempts to address the gap by modelling the influence of personal and household socio-demographic attributes, the built environment, commute characteristics, and personal attitudes of the students on the monthly usage frequency of five different types of commute modes (auto, transit, ride-hailing, bicycle, and walk). It uses data from a large-scale student travel survey conducted among 10 post-secondary institutions in the Greater Toronto and Hamilton Area. The study makes use of a sophisticated modelling approach, consisting of a multiple indicators multiple causes model and a zero-inflated ordered probit model to analyze the factors affecting the decision to use and the usage frequency of the exhaustive set of commute modes. The findings emphasize the importance of commute distance, available resources (in terms of mobility tools, living situation, household vehicles, and income), and built environment attributes in the usage frequency of different commute modes. Marginal effects are used to inform actionable policy recommendations for both the institutions and the regional municipalities. The recommendations include offering discounted and promotional transit passes to encourage students to use public transit frequently, increasing the capacity of student housing and enhancing the sidewalk and bicycle infrastructure within 3–5 km of the campus locations, and increasing the transit accessibility of the institutions by establishing subway stops in proximity to the campuses. These recommendations, when implemented, will help to adequately meet the travel needs of the students while also improving their overall campus life experience.

Analysis of the injury-severity outcomes of maritime accidents using a zero-inflated ordered probit model

This study provides an empirical analysis of the injury severity outcomes of maritime accidents by exploring the influential factors for two underlying injury severity states: the injury-free state and the injury-prone state. The former may reflect the generation mechanism of accidents with limited potential to result in injury outcomes, whereas, the latter may represent injury severity when the accident falls into the injury-prone category. To account for the possible presence of these two underlying regimes, a zero-inflated ordered probit (ZIOP) model is employed using injury-severity data extracted from 1,128 maritime accident investigation reports between 2000 and 2019. The results indicate that on one hand, capsizing/sinking, hull/machinery damage and other accident type, adverse sea state, poor education background and short period of holding the present rank are more likely to be injury-prone. On the other hand, gross tonnage and water depth, distance from the coast, flag state of convenience, and accident type impact the likelihood of severe injuries if the accident is in the injury-prone category. The marginal effect analysis highlights some interesting effects caused by sea state, ship manning and gross tonnage, as well as accident type and location. The results of Akaike's information criteria (AIC), Bayesian information criteria (BIC) and Vuong's test show that the ZIOP model outperforms the traditional ordered probit model and can serve as an alternative to study the injury severity of maritime accidents.

Smoking Behavior Among Students: Using Health Belief Model and Zero-Inflated Ordered Probit Model

Background: Smoking is increasing among adolescents and young adults. Adolescents’ smoking can predict frequent smoking in early adulthood. This study aimed to explore the predictors of smoking among high school students using health belief model (HBM). Materials and Methods: In this cross-sectional study conducted in 2019 in Bandar-Abbas city, South of Iran, 444 male high-school students aged 15-19 years (mean age: 16.7±0.85) were explored. Data were collected using a researcher-made questionnaire consisted of two main sections: sociodemographic characteristics and HBM constructs. The zero-inflated ordered probit (ZIOP) model was adopted for investigating the association between HBM constructs and smoking behavior. Results: Results indicated that 82% of the participants never smoked cigarette. The results of ZIOP model showed that the knowledge (P=0.026), susceptibility (P<0.001), severity (P=0.035), benefits (P=0.004), and cues to action (P=0.019) had significant effects on smoking cigarette after being adjusted for other covariates (i.e., age, parents’ education, losing one of the parents). Moreover, having a smoker friend was found to be an inflation factor (P<0.001). Adolescents with smoking friends were 44% less likely to avoid smoking. Conclusion: It was concluded that having a smoker friend, knowledge, susceptibility, severity, benefits, and cues to action had considerable predictive capacity for predicting smoking attitude. Therefore, it was recommended that these factors should be seriously considered when designing educational programs with the aim of reducing adolescent smoking.

Examining the influence of attitudinal factors on the use of ride-hailing services in Toronto

The continued growth of ride-hailing usage creates the need for policymakers to understand the factors that affect the adoption and utilization of ride-hailing services. Attitudinal and perceptual factors are of particular importance, both because ride-hailing services are still evolving, and a relatively small number of studies have examined the role of these factors. This paper utilizes data from a web-based survey to understand the role that latent attitudinal factors play in adopting and using ride-hailing services in Toronto. Specifically, two binary logistic regression models are used to understand the factors that influence the adoption of exclusive and shared ride-hailing services. Besides, a zero-inflated ordered probit (ZIOP) model is estimated to investigate the factors that affect the frequency with which a person uses ride-hailing. The empirical investigation reveals that the perception of ride-hailing services tends to differ between individuals with ride-hailing experience and those without, which is expected given the relative novelty of ride-hailing. The logistic regression models reveal that, although common attributes affect the likelihood that a person has adopted a ride-hailing service, the influence of these factors varies based on the specific type of service. This underscores the value of distinguishing between exclusive and shared ride-hailing services. The ZIOP model shows that attitudinal factors regarding qualitative trip characteristics, the inclination towards using ride-hailing services in certain situations, and the consideration of parking requirements affect the frequency with which a person uses ride-hailing. Also, transit pass ownership was found to influence the frequency with which a person uses ride-hailing positively. The results of this study aim to provide insights into the adoption and utilization of ride-hailing, which can help inform policies that aim to encourage the use of shared ride-hailing as an alternative to exclusive ride-hailing services.

Estimation of nested and zero-inflated ordered probit models

We introduce three new commands—nop, ziop2, and ziop3—for the estimation of a three-part nested ordered probit model, the two-part zero-inflated ordered probit models of Harris and Zhao (2007, Journal of Econometrics 141: 1073–1099) and Brooks, Harris, and Spencer (2012, Economics Letters 117: 683–686), and a three-part zero-inflated ordered probit model of Sirchenko (2020, Studies in Nonlinear Dynamics and Econometrics 24: 1) for ordinal outcomes, with both exogenous and endogenous switching. The three-part models allow the probabilities of positive, neutral (zero), and negative outcomes to be generated by distinct processes. The zero-inflated models address a preponderance of zeros and allow them to emerge in different latent regimes. We provide postestimation commands to compute probabilistic predictions and various measures of their accuracy, to assess the goodness of fit, and to perform model comparison using the Vuong test (Vuong, 1989, Econometrica 57: 307–333) with the corrections based on the Akaike and Schwarz information criteria. We investigate the finite-sample performance of the maximum likelihood estimators by Monte Carlo simulations, discuss the relations among the models, and illustrate the new commands with an empirical application to the U.S. federal funds rate target.

A Zero-Inflated Ordered Probit Model to Analyze Hazmat Truck Drivers’ Violation Behavior and Associated Risk Factors

There are few studies on the violation of truck drivers, especially the hazmat truck driver, although truck driver's violation may cause serious casualties. This paper aims to investigate hazmat truck drivers' violation behavior and identify associated risk factors. Different data sources in intelligent transportation system (ITS) including hazmat transportation management system and traffic safety management system are extracted and emerged together. Three years (2016-2018) of violation data that comprised 11612 trip record in China are employed in this research. Based on Bayesian theory, this study proposes zero-inflated ordered probit (ZIOP) model and three alternative models to exploring the relationship between hazmat truck drivers' violation frequency and the key risk factors. The results show that ZIOP model can handle excessive zero observation problem of violation data properly and differentiate between `always-zero group' drivers and drivers who did not violate the traffic rules during research period but would do so in different surroundings. The results also indicate that the violation probability and the violation frequency level of hazmat truck drivers are influenced by driver characteristics, freight order attributes, and drivers' violation records. This research provides guidance for driving training and safety education of hazmat truck drivers, and will be helpful in building better driving simulation models.

Incorporating car owner preferences for the introduction of economic incentives for speed limit enforcement

Human error including driving misbehavior contributes to over 90 percent of road vehicle accidents, and speeding is considered to be risky. Smart technologies, such as Connected Vehicle System (CVS) are among the interesting technical options to improve driving behavior, and Pay-As-You-Speed (PAYS) is an effective economic incentive to reduce speed violations. We investigated the acceptability of CVS with and without the presence of economic incentives, such as PAYS, in the context of a middle-income country: Iran. We used a Zero-Inflated Ordered Probit model (ZIOP) to estimate drivers’ willingness to pay for a CVS, and a hazard-based model for predicting the incentive level needed for accepting CVS via a PAYS scheme. ZIOP model indicated that drivers with the following characteristics were more likely to pay more for CVS: having a comprehensive insurance coverage, being younger than 60 years, owning more than one car, and having older vehicles. The hazard-based model also confirmed that drivers that speed relatively often have a lower tendency to adopt CVS, and drivers who experienced an accident in the past were more inclined to adopt CVS via PAYS. Also, drivers' opinion about CVS, vehicle characteristics, demographics, and driving experience influenced the effect of PAYS characteristics on acceptability of CVS. Finally, we offer recommendations for how to effectively implement CVS, in order to significantly reduce the high fatality and accident rates in middle-income countries such as Iran.

Active travel for active ageing in China: The role of built environment

China has been witnessing prominent demographic ageing because of its sustained low fertility (one-child policy) and mortality rates. In 2017, nearly one in four elderly adults in the world live in China. The rapid increase of the elderly population is supposed to dramatically influence the urban and transportation system. Active travel plays an important role for the ageing Chinese population to sustain their mobility and wellbeing. To provide suitable policy implications for age-friendly travel environments in China, this study investigates how the built environment affects active travel behavior. Particularly, we explore the influences of built environment on daily active travel frequency and time expenditure while taking into account travel attitudes. A zero-inflated ordered probit model and a Cox proportional hazards model are respectively estimated based on the Nanjing Travel Survey data. Results show that the social and cultural contexts exert pronounced impacts on the travel pattern of Chinese older people. Specifically, it is found that the living pattern of co-residence, and the proximity to market, park/square, and chess/card room are influential in shaping active travel patterns. In addition, the built environment shows larger effects on the active travel behavior of older adults than on that of young people. Attitudes towards active travel are not significant in explaining the senior's travel behavior, indicating limited self-selection effects. The findings will offer insights to establish effective and appropriate land use strategies and public facility distribution for the elderly during the Chinese urban renewal process.

Investigating the factors affecting secondary crash frequency caused by one primary crash using zero-inflated ordered probit regression

The contributing factors to secondary crashes have been investigated by a number of studies in recent years. However, previous studies generally considered only the first secondary crash after the primary crash. Existing studies have seldom considered the multiple secondary crashes caused by one primary crash and the effects of real-time traffic flow conditions. This paper aimed to investigate the effects of real-time traffic flow conditions on the frequency of secondary crashes caused by one primary crash on freeways. The zero-inflated ordered probit (ZIOP) regression model was developed to link the probability of multiple secondary crashes after the occurrence of one primary crash with real-time traffic flow, geometric, weather and primary crash characteristics. The ZIOP regression model analyzed the probability of secondary crash frequency after one primary crash by separating it into two states. One is a secondary-crash-free state that determines whether the occurrence of a crash will lead to one or more secondary crashes, and the other is a secondary-crash-prone state that determines the secondary crash frequency caused by one primary crash. The average speed, average traffic volume, and the difference between the numbers of on-ramp and off-ramp are the significant variables in the secondary-crash-free state. In the secondary-crash-prone state, the significant variables affecting the probability of multiple secondary crashes include average detector occupancy, rainy weather, primary crash severity, and hit-and-run primary crash. The ROC curves were used to test predictive performance of the ZIOP model. The test results suggested that the ZIOP model provide reasonably good predictive accuracy of multiple secondary crashes caused by one primary crash.

What drives the use of ridehailing in California? Ordered probit models of the usage frequency of Uber and Lyft

The availability of ridehailing services, such as those provided by Uber and Lyft in the U.S. market, as well as the share of trips made by these services, are continuously growing. Yet, the factors affecting the frequency of use of these services are not well understood. In this paper, we investigate how the frequency of use of ridehailing varies across segments of the California population and under various circumstances. We analyze data from the California Millennials Dataset (N = 1975), collected in fall 2015 through an online survey administered to both millennials and members of the preceding Generation X. We estimate an ordered probit model with sample selection and a zero-inflated ordered probit model with correlated error terms to distinguish the factors affecting the frequency of use of ridehailing from those affecting the adoption of these services. The results are consistent across models: sociodemographic variables are important predictors of service adoption but do not explain much of the variation in the frequency of use. Land use mix and activity density respectively decrease and increase the frequency of ridehailing. The results also confirm that individuals who frequently use smartphone apps to manage other aspects of their travel (e.g. to select a route or check traffic) are more likely to adopt ridehailing and use it more often. This is also true for long-distance travelers, in particular, those who frequently travel by plane for leisure purposes. Individuals with higher willingness to pay to reduce their travel time use ridehailing more often. Those with stronger preferences to own a personal vehicle and those with stronger concerns about the safety/security of ridehailing are less likely to be frequent users. These results provide new insights into the adoption and use of ridehailing that could help to inform planning and forecasting efforts.

French Arms Exports and Intrastate Conflicts: An Empirical Investigation

The aim of this article is to evaluate how French Major Conventional Weapons (MCW) exports impact on the conflict intensity of recipient countries. The recent increase in French arms exports seems to contradict the French political discourse on the promotion of regional stability. We run zero-inflated ordered probit model in order to analyze the role of the arms trade on the intensity of civil conflicts in 144 countries from 1992 to 2014, using SIPRI and UCDP/PRIO data. Our results suggest that French MCW exports tended not to exacerbate intrastate conflicts during this period. This finding is robust to changes in the empirical framework. We propose two lines of explanations: France seems to be prone to choosing partners that respect human rights and selling more ‘defense-oriented’ MCW than the rest of the world.

Estimation of Nested and Zero-Inflated Ordered Probit Models

We introduce three new STATA commands, nop, ziop2 and ziop3, for the estimation of a three-part nested ordered probit model, the two-part zero-inflated ordered probit models of Harris and Zhao (2007, Journal of Econometrics 141: 1073-1099) and Brooks, Harris and Spencer (2012, Economics Letters 117: 683-686), and a three-part zero-inflated ordered probit model for ordinal outcomes, with both exogenous and endogenous switching. The three-part models allow the probabilities of positive, neutral (zero) and negative outcomes to be generated by distinct processes. The zero-inflated models address a preponderance of zeros and allow them to emerge in different latent regimes. We provide postestimation commands to compute probabilistic predictions and various measures of their accuracy, to access the goodness of fit, and to perform model comparison using the Vuong test (Vuong 1989, Econometrica 57: 307-333) with the corrections based on the Akaike and Schwarz information criteria. We investigate the finite-sample performance of the maximum likelihood estimators by Monte Carlo simulations, discuss the relations among the models, and illustrate the new commands with an empirical application to the U.S. federal funds rate target.

Zero-inflated ordered probit approach to modeling mushroom consumption in the United States

This paper investigates the determinants of fresh and processed mushroom consumption in the United States by employing the zero-inflated ordered probit (ZIOP) model. The ZIOP model accounts for excessive zero observations and allows us to differentiate between genuine non-consumers and individuals who did not consume during the given period but might under different circumstances. The results indicate that the market for fresh mushrooms is larger than that for processed mushrooms. However, the market for processed mushrooms has a larger portion of potential consumers which might indicate more potential if appropriate marketing strategies are applied. The results also suggest that the decisions to participate in the market or not and the consumption frequency are driven by structurally different factors. A comparison of the ZIOP to other models is included to show the advantages of allowing for non-consumers and potential consumers to be analyzed separately.

영과잉 순서형 프로빗 모형의 베이지안 추론 : 한국형 베일리-III 검사 자료에의 적용

영과잉 순서형 프로빗 모형의 베이지안 추론 : 한국형 베일리-III 검사 자료에의 적용

Modelling Tobacco Consumption with a Zero-Inflated Ordered Probit Model ∗

Data for discrete ordered random variables are often characterised by “excessive” zero observations. Traditional ordered probit models have limited capacity in explaining the preponderance of zero observations, especially when the zeros may relate to two distinct situations of non-participation and infrequent participation (or consumption), for example. We propose a zero-inflated ordered probit (ZIOP) model using a double-hurdle combination of a split (probit) model and an ordered probit model which, potentially, relate to different sets of covariates. Monte Carlo results suggest that the new model performs well. Finally, the model is applied to a consumer choice problem of tobacco consumption.

Cigarette consumption by individuals in Malaysia: a zero-inflated ordered probability approach

To investigate cigarette consumption patterns exhibited by non-smokers, occasional smokers, and daily smokers in Malaysia. A sample of 4204 individuals from the 2011 Malaysian Global Adult Tobacco Survey is analyzed. A zero-inflated ordered probit model is used to accommodate the ordinal nature of smoking outcomes with excessive zero observations of non-smokers. Socio-demographic characteristics are closely associated with consumption patterns of non-smokers, occasional smokers, and daily smokers. Specifically, urbanites, government employees, and Malays exhibit greater tendencies to be daily smokers and lower propensities to be non-smokers than others. Education is a deterring factor in cigarette smoking as higher education up to the tertiary level raises the propensity of being a non-smoker by 8.16 percentage points. The role of ethnicity is highlighted as individuals of Malay and other ethnic backgrounds are more likely to smoke daily, while exhibiting lower propensities of being non-smokers. Males are more likely to engage in occasional or daily smoking than females. Employment in government or non-government sectors increases the probability of daily smoking. Our results suggest the need to include measures to cope with internal or external cues among smokers with specific socio-demographic characteristics. Cessation interventions should focus on daily smokers in urban surroundings with low education levels and those of Malay ethnic origins. Anti-smoking measures in line with occasional smoking may include ameliorating the male attitude toward smoking and policies to prohibit workplace indulgence.

A categorical modeling approach to analyzing the impacts of the Lacey Act 2008 amendment on Chinese companies’ export cost and the implications on their sourcing behaviors

The United States (US) Lacey Act 2008 amendment (LAA) is a timber legality regulation that requires US importers to monitor and minimize the risk of illegally harvested wood products within their supply chains. This paper empirically examines the effect of the LAA on Chinese companies’ export costs to the US. The study uses 138 responses from two surveys in Shanghai, China, in 2013, i.e., 5 years after the LAA was implemented. Given the high proportion of zero export increase indicated by the Chinese companies, a zero-inflated ordered probit model was used to model Chinese companies’ export cost increases to the US. The research results demonstrate that pre-LAA raw material sourcing patterns are primary indicators of the respondents’ export cost increase to the US as a result of the LAA. From the results, it can be inferred that log and lumber importers from suspect regions are taking additional measures, by changing their procurement practices, to ensure the legality of their raw material, which is adding to their cost structure. The results also indicate that smaller companies, given their flexibility with raw material procurement, were less likely to experience a post-LAA cost increase relative to their larger counterparts.