Effect Of Unobserved Heterogeneity Research Articles

Job mismatches and career mobility

ABSTRACTDoes over-education assist or hinder occupational advancement? Career mobility theory hypothesizes that over-education leads to a higher level of occupational advancement and wage growth over time, with mixed international empirical evidence. This paper re-tests career mobility theory directly using a rich Australian longitudinal data set. A dynamic random effects probit model is employed to examine upward occupational mobility, considering two-digit occupational rank advancement and wage growth over three-year intervals. The ‘Household, Income and Labour Dynamics in Australia’ data across nine years are employed, and a Mundlak correction model is adopted to adjust for unobserved heterogeneity effects and potential endogeneity, both of which are important to over-education analysis. Contrary to career theory, the results point to job mismatch as an economic concern rather than a passing phase, regardless of whether or not workers are skill-matched. Results further show the importance of adjusting for endogeneity.

IFRS Adoption and Audit Delay: The Role of Shareholders in the Audit Committee

This study examined the relationship between audit delay after IFRS adoption and the role of shareholders in the audit committee as well as testing the difference of pre-and post IFRS adoption periods. A sample of 101 firms with 505 firm-year observations over five year period for firms listed on the Nigeria Stock Exchange was employed for the study, utilizing data from the annual report and accounts of the sample firms. Generalized Methods of Moment (GMM) estimation was used to check the effects of unobserved heterogeneity in audit delay model, while the test of difference in R2 value for pre-and post-adoption periods was determined using Cramer’s Z-statistics. Findings indicate that audit report lag is faster with shareholders in the audit committee. The study proved that brand named auditors such as Big4 can significantly perform faster audit task than non-Big4 firms in IFRS regime. The importance of the study’s findings demonstrates statistical inference on value relevance increase based on the unique IFRS adoption in Nigeria. Thus, regulators should consider increasing the tenure of shareholders in the audit committee to enable them to become more familiar with the corporate reporting under IFRS regime.

Exploring unobserved heterogeneity in bicyclists’ red-light running behaviors at different crossing facilities

Bicyclists running the red light at crossing facilities increase the potential of colliding with motor vehicles. Exploring the contributing factors could improve the prediction of running red-light probability and develop countermeasures to reduce such behaviors. However, individuals could have unobserved heterogeneities in running a red light, which make the accurate prediction more challenging. Traditional models assume that factor parameters are fixed and cannot capture the varying impacts on red-light running behaviors. In this study, we employed the full Bayesian random parameters logistic regression approach to account for the unobserved heterogeneous effects. Two types of crossing facilities were considered which were the signalized intersection crosswalks and the road segment crosswalks. Electric and conventional bikes were distinguished in the modeling. Data were collected from 16 crosswalks in urban area of Nanjing, China. Factors such as individual characteristics, road geometric design, environmental features, and traffic variables were examined. Model comparison indicates that the full Bayesian random parameters logistic regression approach is statistically superior to the standard logistic regression model. More red-light runners are predicted at signalized intersection crosswalks than at road segment crosswalks. Factors affecting red-light running behaviors are gender, age, bike type, road width, presence of raised median, separation width, signal type, green ratio, bike and vehicle volume, and average vehicle speed. Factors associated with the unobserved heterogeneity are gender, bike type, signal type, separation width, and bike volume.

Modeling Fiscal Sustainability in Dynamic Macro-Panels with Heterogeneous Effects: Evidence from German Federal States

In this paper, we extend Henning Bohn’s (2008) fiscal sustainability test by allowing for slope heterogeneity and cross-sectional dependence (CD). In particular, our econometric approach is the first that allows fiscal reaction functions (FRF) to capture unobserved heterogeneous effects from business and fiscal policy cycles. We apply this econometric approach to sub-national public finance data of the German Laender between 1950 and 2015 and find that their fiscal policy only partly meets fiscal sustainability criteria. According to our results, politicians have significantly reacted to increasing debt levels by increasing budget surpluses since 1991. However, time-series evidence for longer periods does not indicate a significant and positive reaction to increasing debt levels in the West German Laender panel.

Interacting effects of unobserved heterogeneity and individual stochasticity in the life history of the southern fulmar.

Individuals are heterogeneous in many ways. Some of these differences are incorporated as individual states (e.g. age, size, breeding status) in population models. However, substantial amounts of heterogeneity may remain unaccounted for, due to unmeasurable genetic, maternal or environmental factors.Such unobserved heterogeneity (UH) affects the behaviour of heterogeneous cohorts via intra‐cohort selection and contributes to inter‐individual variance in demographic outcomes such as longevity and lifetime reproduction. Variance is also produced by individual stochasticity, due to random events in the life cycle of wild organisms, yet no study thus far has attempted to decompose the variance in demographic outcomes into contributions from UH and individual stochasticity for an animal population in the wild.We developed a stage‐classified matrix population model for the southern fulmar breeding on Ile des Pétrels, Antarctica. We applied multievent, multistate mark–recapture methods to estimate a finite mixture model accounting for UH in all vital rates and Markov chain methods to calculate demographic outcomes. Finally, we partitioned the variance in demographic outcomes into contributions from UH and individual stochasticity.We identify three UH groups, differing substantially in longevity, lifetime reproductive output, age at first reproduction and in the proportion of the life spent in each reproductive state.–14% of individuals at fledging have a delayed but high probability of recruitment and extended reproductive life span.–67% of individuals are less likely to reach adulthood, recruit late and skip breeding often but have the highest adult survival rate.–19% of individuals recruit early and attempt to breed often. They are likely to raise their offspring successfully, but experience a relatively short life span. Unobserved heterogeneity only explains a small fraction of the variances in longevity (5.9%), age at first reproduction (3.7%) and lifetime reproduction (22%).UH can affect the entire life cycle, including survival, development and reproductive rates, with consequences over the lifetime of individuals and impacts on cohort dynamics. The respective role of UH vs. individual stochasticity varies greatly among demographic outcomes. We discuss the implication of our finding for the gradient of life‐history strategies observed among species and argue that individual differences should be accounted for in demographic studies of wild populations.

Breast cancer screening among women in Namibia: explaining the effect of health insurance coverage and access to information on screening behaviours.

Breast cancer contributes substantially to morbidity and mortality in Namibia as is the case in most countries in Sub-Saharan Africa (SSA). However, there is a dearth of nationally representative studies that examine the odds of screening for breast cancer in Namibia and SSA at large. This paper aims to fill this gap by examining the determinants of breast cancer screening guided by the Health Belief Model. We applied hierarchical binary logit regression models to explore the determinants of breast cancer screening using the 2013 Namibia Demography and Health Survey (NDHS). We accounted for the effect of unobserved heterogeneity that may affect breast cancer, testing behaviours among women cluster level. The NDHS is a nationally representative dataset that has recently started to collect information on cancer screening. The results show that women who have health insurance coverage (odds ratio (OR) = 1.62, p ≤ 0.01), maintain contact with health professionals (OR = 1.47, p = 0.01), and who have secondary (OR = 1.38, p = 0.01) and higher (OR = 1.77, p ≤ 0.01) education were more likely to be screened for breast cancer. Factors that influence women's perception of their susceptibility to breast cancer such as birthing experience, age, region and place of residence were associated with screening in this context. Overall, the health belief model predicted women's testing behaviours and also revealed the absence of relevant risk factors in the NDHS data that might influence screening. Overall, our results show that strategies for early diagnosis of breast cancer should be given major priority by cancer control boards as well as ministries of health in SSA. These strategies should centre on early screening and may involve reducing or eliminating barriers to health care, access to relevant health information and encouraging breast self-examination.

Estimating Input Complementarities with Unobserved Heterogeneity: Evidence from Ethiopia

AbstractThe level of technology adoption is often characterised as low in Africa. Recent evidence, however, points to the coexistence of substantial heterogeneity across farm households and the lack of a suitable mix of inputs for farmers to take advantage of input complementarities. We use a random coefficients multivariate probit model to quantify the complementarities between agricultural inputs and alternative forms of unobserved heterogeneity effects in modeling farmers' technology adoption decisions. The empirical analysis reveals that, conditional on various types of unobserved heterogeneity effects, farmers' technology adoption decisions exhibit strong complementarity for some inputs. The analysis also reveals substantial unobserved heterogeneity effects. We show that ignoring these behavioural features (unobserved heterogeneity and input complementarity) has important implications in quantifying the effect of some policy interventions that are meant to facilitate technology adoption. In particular, ignoring these features leads to significant overestimation of the effectiveness of extension services.

Occupant-level injury severity analyses for taxis in Hong Kong: A Bayesian space-time logistic model

This study aimed to identify the factors affecting the crash-related severity level of injuries in taxis and quantify the associations between these factors and taxi occupant injury severity. Casualties resulting from taxi crashes from 2004 to 2013 in Hong Kong were divided into four categories: taxi drivers, taxi passengers, private car drivers and private car passengers. To avoid any biased interpretation caused by unobserved spatial and temporal effects, a Bayesian hierarchical logistic modeling approach with conditional autoregressive priors was applied, and four different model forms were tested. For taxi drivers and passengers, the model with space-time interaction was proven to most properly address the unobserved heterogeneity effects. The results indicated that time of week, number of vehicles involved, weather, point of impact and driver age were closely associated with taxi drivers’ injury severity level in a crash. For taxi passengers’ injury severity an additional factor, taxi service area, was influential. To investigate the differences between taxis and other traffic, similar models were established for private car drivers and passengers. The results revealed that although location in the network and driver gender significantly influenced private car drivers’ injury severity, they did not influence taxi drivers’ injury severity. Compared with taxi passengers, the injury severity of private car passengers was more sensitive to average speed and whether seat belts were worn. Older drivers, urban taxis and fatigued driving were identified as factors that increased taxi occupant injury severity in Hong Kong.

Modeling single-vehicle run-off-road crash severity in rural areas: Accounting for unobserved heterogeneity and age difference

This study investigates factors that significantly contribute to the severity of driver injuries resulting from single-vehicle run-off-road (SV ROR) crashes. A mixed logit model approach is employed to explore the potential unobserved heterogeneous effects associated with each age group: young (ages 16–24), middle-aged (ages 25–65), and older drivers (ages over 65). Likelihood ratio tests indicated that the development of separate injury severity models for each age group is statistically superior to estimating a single model using all data. Based on the crash data collected from 2009 to 2013 in North Carolina, a series of driver, vehicle, roadway, and environmental characteristics are examined. Both parameter estimates and their elasticities are developed and used to interpret the models. The estimation results show that contributing factors which significantly affect the injury severity of an SV ROR crash differ across three age groups. Use of restraint device and horizontal curves are found to affect crash injuries and fatalities in all age groups. Reckless driving, speeding, distraction, inexperience, drug or alcohol involvement, presence of passengers, and driving an SUV or a van are found to have a more pronounced influence in young and middle-aged drivers than older drivers. Compared to the passenger cars, older drivers are less likely to experience possible injuries in a large-size vehicle (e.g., truck or bus). The average annual daily traffic volume and lighting conditions are also found to influence the resulting injury severity of an SV ROR crash specific to young drivers.

The Influences of Health Insurance and Access to Information on Prostate Cancer Screening among Men in Dominican Republic.

Objectives. Although research demonstrates the public health burden of prostate cancer among men in the Caribbean, relatively little is known about the factors that underlie the low levels of testing for the disease among this population. Study Design. A cross-sectional study of prostate cancer testing behaviours among men aged 40–60 years in Dominican Republic using the Demographic and Health Survey (2013). Methods. We use hierarchical binary logit regression models and average treatment effects combined with propensity score matching to explore the determinants of prostate screening as well as the average effect of health insurance coverage on screening. The use of hierarchical binary logit regression enabled us to control for the effect of unobserved heterogeneity at the cluster level that may affect prostate cancer testing behaviours. Results. Screening varied significantly with health insurance coverage, knowledge of cholesterol level, education, and wealth. Insured men were more likely to test for prostate cancer (OR = 1.65, p = 0.01) compared to the uninsured. Conclusions. The expansion and restructuring of Dominican Republic universal health insurance scheme to ensure equity in access may improve health access that would potentially impact positively on prostate cancer screening among men.

What Prevents Men Aged 40-64 Years from Prostate Cancer Screening in Namibia?

Objectives. Although a growing body of evidence demonstrates the public health burden of prostate cancer in SSA, relatively little is known about the underlying factors surrounding the low levels of testing for the disease in the context of this region. Using Namibia Demographic Health Survey dataset (NDHS, 2013), we examined the factors that influence men's decision to screen for prostate cancer in Namibia. Methods. We use complementary log-log regression models to explore the determinants of screening for prostate cancer. We also corrected for the effect of unobserved heterogeneity that may affect screening behaviours at the cluster level. Results. The results show that health insurance coverage (OR = 2.95, p = 0.01) is an important predictor of screening for prostate cancer in Namibia. In addition, higher education and discussing reproductive issues with a health worker (OR = 2.02, p = 0.05) were more likely to screening for prostate cancer. Conclusions. A universal health insurance scheme may be necessary to increase uptake of prostate cancer screening. However it needs to be acknowledged that expanded screening can have negative consequences and any allocation of scarce resources towards screening must be guided by evidence obtained from the local context about the costs and benefits of screening.

A note on generalized ordered outcome models

While there is growing application of generalized ordered outcome model variants (widely known as Generalized Ordered Logit (GOL) model and Partial Proportional Odds Logit (PPO) model) in crash injury severity analysis, there are several aspects of these approaches that are not well documented in extant safety literature. The current research note presents the relationship between these two variants of generalized ordered outcome models and elaborates on model interpretation issues. While these variants arise from different mathematical approaches employed to enhance the traditional ordered outcome model, we establish that these are mathematically identical. We also discuss how one can facilitate estimation and interpretation while building on the ordered outcome model estimates – a useful process for practitioners considering upgrading their existing traditional ordered logit/probit injury severity models. Finally, the note presents the differences within GOL and PPO model frameworks, for accommodating the effect of unobserved heterogeneity, referred to as Mixed Generalized Ordered Logit (MGOL) and Mixed Partial Proportional Odds Logit (MPPO) models while also discussing the computational difficulties that may arise in estimating these models.

Simultaneous equation modeling of freeway accident duration and lanes blocked

Unbiased estimates are crucial to providing correct information that is required in accident management. Whereas earlier studies have proposed various approaches to reduce the adverse effects of unobserved heterogeneity, this study proposes accelerated failure time (AFT) models of accident duration with flexible distributions of the error term and the frailty parameter. A two-stage residual inclusion (2SRI) method for identifying the effects of endogenous variable in the nonlinear AFT models is adopted. The Taiwanese freeway accident databases were used to evaluate the applicability of the proposed approach. The results reveal that AFT models with a non-monotone Log-logistic error term distribution and a gamma frailty parameter distribution outperformed those with other assumed distributions. Number of lanes blocked was found as a significant endogenous variable to accident duration; the endogenous variable effect, if not accounted for or accounted by traditional two-stage least squares estimators, would underestimate the effect of the number of lanes blocked on accident duration.

Investigating the disparities in cervical cancer screening among Namibian women

ObjectivesWe examined the influence of knowledge and information, health care access and different socio-economic variables on women's decision to screen for cervical cancer using a nationally representative dataset. MethodsWe use hierarchical binary logit regression models to explore the determinants of screening for cervical cancer among women who reported hearing about cervical cancer. This enabled us to include the effect of unobserved heterogeneity at the cluster level that may affect screening behaviors. ResultsAmong women who have heard about cervical cancer (N=6542), only 39% of them did undergo screening with a mean age of 33years. The univariate results reveal that women who are educated, insured, can afford money needed for treatment and reported distance not a barrier to accessing healthcare were more likely to screen. Our multivariate results indicate that insured women (OR=1.89, p=0.001) and women who had access to information through education and contact with a health worker (OR=1.41, p=0.001) were more likely to undertake screening compared to uninsured women and those with no contact with a health personnel, after controlling for relevant variables. ConclusionsThe adoption of a universal health insurance scheme that ensures equity in access to health care and extension of public health information targeting women in rural communities especially within the Caprivi region may be needed for a large scale increase in cervical cancer screening in Namibia.

How to deal with heterogeneity among tourism constructs?

•Analyzing the effects of unobserved heterogeneity in Structural Equation Modeling.•Uncovering unobserved heterogeneity and segments by a finite-mixture approach.•Examining how environmental attitudes influence the preference for outdoor activity.•Model estimation in a nature-based tourism context using PLS Path Modeling.

Production Risks and Food Security under Alternative Technology Choices in Malawi: Application of a Multinomial Endogenous Switching Regression

AbstractEmploying nationally representative data, we investigate the impact of Sustainable Intensification Practices (SIPs) on farm households’ food security, downside risk and the cost of risk in Malawi. The analysis relies on a flexible moment‐based specification of a stochastic production function in a multinomial endogenous switching regression framework to correct for the selection bias stemming both from observed and unobserved heterogeneity. A quantile moment approach is used to estimate the cost of risk. After controlling for the effects of unobserved heterogeneity and several observable variables on maize production and downside risk functions, estimation results show that the adoption ofSIPs increases food security and reduces downside risk exposure and the cost of risk. We estimate greater food security and larger reduction in downside risk from simultaneous adoption of both crop diversification (maize–legume intercropping and rotations) and minimum tillage, suggesting that there are complementary benefits from these practices. We find most of the cost of risk comes from exposure to downside risk. Our findings imply that in dealing with production risks development agents should encourage the adoption of agronomic and resource‐management practices along with other risk mitigation and food security improving strategies.

The joint analysis of injury severity of drivers in two-vehicle crashes accommodating seat belt use endogeneity

The current study contributes to the existing injury severity modeling literature by developing a multivariate probit model of injury severity and seat belt use decisions of both drivers involved in two-vehicle crashes. The modeling approach enables the joint modeling of the injury severity of multiple individuals involved in a crash, while also recognizing the endogeneity of seat belt use in predicting injury severity levels as well as accommodating unobserved heterogeneity in the effects of variables. The proposed model is applied to analyze the injury severity of drivers involved in two-vehicle road crashes in Denmark.The empirical analysis provides strong support for the notion that people offset the restraint benefits of seat belt use by driving more aggressively. Also, men and those individuals driving heavy vehicles have a lower injury risk than women and those driving lighter vehicles, respectively. At the same time, men and individuals driving heavy vehicles pose more of a danger to other drivers on the roadway when involved in a crash. Other important determinants of injury severity include speed limit on roadways where crash occurs, the presence (or absence) of center dividers (median barriers), and whether the crash involves a head-on collision. These and other results are discussed, along with implications for countermeasures to reduce injury severities in crashes. The analysis also underscores the importance of considering injury severity at a crash level, while accommodating seat belt endogeneity effects and unobserved heterogeneity effects.

Alcohol Consumption and Pregnancies Among Youth: Evidence from a Semi-Parametric Approach

Despite a well-established correlation between alcohol intake and various risk-taking sexual behaviors, the causality remains unknown. The observed association can be easily attributed to the influence of unobserved individual characteristics rather than the influence of substance use. I model the effect of alcohol use on the likelihood of pregnancy among youth using a variety of estimation techniques. The preference is given to the semi-parametric model where the cumulative distribution of heterogeneity is approximated by a 4-point discrete distribution. Using data on 17-28 year-old women from the National Longitudinal Survey of Youth, I find that alcohol consumption increases the likelihood of pregnancy by 4.7 percentage points. Quantitatively similar but statistically weaker effects were found in the fully parametric models such as the two-stage least squares model and the bivariate probit model. Finally, the fully parametric models that ignore the effect of unobserved heterogeneity failed to establish this relationship.

She’s got a ticket to ride: gender and public transit passes

Drawing on a household survey spanning 1997–2007 from Germany, this analysis investigates the decision to purchase a weekly or monthly transit pass against the backdrop of two questions: (1) Does gender play a role in determining the probability that an individual owns a pass? (2) If so, how is this role mitigated or exacerbated by other socioeconomic attributes of the individual and the household in which they reside? These questions are pursued through a combination of descriptive analyses and econometric methods, the latter of which relies on variants of the probit- and heteroskedastic probit model to control for the effects of unobserved heterogeneity that could otherwise induce biased estimates. The model uncovers several determinants of transit pass patronage over which policy makers have direct leverage, including fuel prices, the siting of transit stops, and the density of transit service. Moreover, while women are found to have a higher probability of owning a transit pass than men, the model identifies few variables whose impact differs according to gender, with the two exceptions being the number of children in the household and the distance to work. This absence of differential effects implies that policy measures to increase ridership are likely to have a roughly uniform impact among women and men.

A competing hazard model of household vehicle transaction behavior with discrete time intervals and unobserved heterogeneity

The analysis of household vehicle ownership behavior is considered as one of the most important elements of household travel behavior studies. Several different modeling frameworks have been employed to develop vehicle ownership models. Among these models, applications of hazard based duration models have drawn particular attention in recent years. Hazard models of vehicle transactions often have overlooked the effects of unobserved heterogeneity in the model. This study attempts to address this limitation by formulating and estimating a discrete competing hazard model of vehicle transaction with multiple outcomes that also accounts for unobserved heterogeneity. Three transaction decisions of purchasing, trading, or disposing of a vehicle are considered and competing duration models, with and without unobserved heterogeneity are estimated and compared. The modeling framework is flexible enough to consider transaction times that are not continuous and the failures that occur within discrete time intervals.