Random Effects Specification Research Articles

Bayesian bivariate Conway–Maxwell–Poisson regression model for correlated count data in sports

Abstract Count data play a crucial role in sports analytics, providing valuable insights into various aspects of the game. Models that accurately capture the characteristics of count data are essential for making reliable inferences. In this paper, we propose the use of the Conway–Maxwell–Poisson (CMP) model for analyzing count data in sports. The CMP model offers flexibility in modeling data with different levels of dispersion. Here we consider a bivariate CMP model that models the potential correlation between home and away scores by incorporating a random effect specification. We illustrate the advantages of the CMP model through simulations. We then analyze data from baseball and soccer games before, during, and after the COVID-19 pandemic. The performance of our proposed CMP model matches or outperforms standard Poisson and Negative Binomial models, providing a good fit and an accurate estimation of the observed effects in count data with any level of dispersion. The results highlight the robustness and flexibility of the CMP model in analyzing count data in sports, making it a suitable default choice for modeling a diverse range of count data types in sports, where the data dispersion may vary.

A spatial panel autoregressive model specification with inverse quantile separation distances of locations

This paper proposes an alternative spatial weight that efficiently captures a spatial dependence. In the past, researchers often used sparse or inverse distance spatial weights. A dense spatial weight is defined by partitioning the separation distances between locations based on quantile values over large spatial scales, where each partition forms conjoint neighborhood sectors and the weights of the respective inverse quantile of separation distances are assigned to each sector. Instead of joint modeling of the spatiotemporal process, a simultaneous spatial panel model is employed after the panel component is imposed on the proposed spatial weight using Kronecker product to perform maximum likelihood estimation and Bayesian inference via MCMC Gibbs method. The specification also involves space, time, and space–time simultaneous components. The performance of the models for the proposed spatial weight is compared with the existing spatial weights using parameter bias. A smaller value of the bias close to zero indicates a stronger value of the spatial error parameter for the proposed spatial weight over the existing spatial weights. It also induces spatial auto-correlations both in the spatial panel data of neighboring locations as well as in the errors. Thus, the proposed method affirms the best efficiency for the dynamic combined spatial panel autoregressive model with random effect specification over the lag and error models, and fixed effect specification.

COVID-19 impact on tourism inflow in selected Asia-Pacific countries: a gravity model framework

PurposeThe main objective of this paper is to examine the impact of COVID-19 on the tourism flows of eight Asia-Pacific Countries: Australia, Hong Kong, Malaysia, New Zealand, the Philippines, Singapore, Taiwan and Thailand.Design/methodology/approachUsing monthly data from 2019M1 to 2021M10 and 48 origin and eight destination countries in a panel Poisson pseudo-maximum likelihood (PPML) estimation technique and gravity equation framework, this paper finds that after controlling for gravity determinants, COVID-19 periods have a 0.689% lower tourism inflow than in non-COVID-19 periods. The total observations in this paper are 12,138.FindingsA 1% increase in COVID-19 transmission in the origin country leads to a 0.037% decline in tourism flow in the destination country, while the reduction is just 0.011% from the destination. On the mortality side, the corresponding decline in tourism flows from origin countries is 0.030%, whereas it is 0.038% from destination countries. A 1% increase in vaccine intensity in the destination country leads to a 0.10% improvement in tourism flows, whereas vaccinations at the source have no statistically significant effect. The results are also robust at a 1% level in a pooled OLS and random-effects specification for the same model.Research limitations/implicationsThe findings provide insights into managing tourism flows concerning transmission, death and vaccination coverage in destination and origin countries.Practical implicationsThe COVID-19-induced tourism decline may also be considered another channel through which the global recession has been aggravated. If we convert this decline in terms of loss of GDP, the global figure will be huge, and airline industries will have to cut down many service products for a long time to recover from the COVID-19-induced tourism decline.Social implicationsIt is to be realized by the policymaker and politicians that infectious diseases have no national boundary, and the problem is not local or national. That’s why it is to be faced globally with cooperation from all the countries.Originality/valueThis is the first paper to address tourism disruption due to COVID-19 in eight Asia-Pacific countries using a gravity model framework.HighlightsAsia-Pacific countries are traditionally globalized through tourism channelsThis pattern was severely affected by COVID-19 transmission and mortality and improved through vaccinationThe gravity model can be used to quantify the loss in the tourism sector due to COVID-19 shocksTransmission and mortality should be controlled both at the origin and the destination countriesVaccinations in destination countries significantly raise tourism flows

Intergenerational transmission of parental risky health behaviors in Chinese children: Are there socioeconomic status differences?

Risky health behaviors in childhood, including smoking, alcohol consumption, and having a poor diet, are the major sources of non-communicable diseases in adulthood. This study aimed to examine how parents affect children's risky health behaviors and whether intergenerational transmission differs based on socioeconomic status (SES). Data were extracted from the 1991-2015 China Health and Nutrition Survey (CHNS). Smoking (n = 5,946), alcohol consumption (n = 7,821), and sugar-sweetened beverages (SSBs) consumption (n = 3,537) were used as proxies for risky health behaviors in children. A binary choice model for panel data with a random-effect specification was employed to examine whether risky health behaviors can be transmitted from parents to their children. Subsequently, we conducted a seemingly unrelated estimation test (SUEST) to explore the differences in parental transmission between the different SES groups. We found strong intergenerational persistence of smoking, alcohol drinking, and SSBs drinking behaviors, except for the mothers' smoking behavior. Mothers had a greater influence on children's alcohol drinking and SSBs drinking behaviors than fathers both in urban and rural areas and in different SES groups. The intergenerational transmission of SSBs drinking behavior exhibited a decreasing trend with increasing SES for both urban and rural families. In urban areas, mothers' alcohol drinking behavior has a decreasing trend with increasing education level, occupation, and income; however, in rural areas, the influence of mothers' alcohol drinking behavior occurred in the same direction with increasing education level and occupation type. In rural areas, the influence of fathers' drinking and smoking behaviors on children appears to mostly increase with increasing SES. Meanwhile, the influence of such behaviors among urban fathers would decrease with increasing SES. Parents' behaviors and SES can influence the initiation of risky health behaviors in their offspring. Thus, to promote healthy behaviors, policymakers can introduce health education programs for parents, particularly for those living in rural areas and with a low SES.

A simple Bayesian state-space approach to the collective risk models

The collective risk model (CRM) for frequency and severity is an important tool for retail insurance ratemaking, natural disaster forecasting, as well as operational risk in banking regulation. This model, initially designed for cross-sectional data, has recently been adapted to a longitudinal context for both a priori and a posteriori ratemaking, through random effects specifications. However, the random effects are usually assumed to be static due to computational concerns, leading to predictive premiums that omit the seniority of the claims. In this paper, we propose a new CRM model with bivariate dynamic random effects processes. The model is based on Bayesian state-space models. It is associated with a simple predictive mean and closed form expression for the likelihood function, while also allowing for the dependence between the frequency and severity components. A real data application for auto insurance is proposed to show the performance of our method.

Probabilistic deterioration modeling of bridge component condition with random effects

ABSTRACT Timely maintenance of bridge components is critical for bridge management functions. With reliable deterioration models, highway agencies can efficiently allocate funding for bridge maintenance and customize maintenance schedules to meet agency budgets. The increased public expectation of acceptable levels of service for bridges coupled with other competing needs makes it crucially important to accurately estimate bridge future conditions so that adequate resources may be allocated for repair and reconstruction purposes. Accurately predicting bridge condition is challenging due to the inherent random nature of factors impacting deterioration, the existence of unobserved variables that are not measured, panel nature of the data and the effects of bridge-specific correlation. Without accounting for these factors, the resulting estimated deterioration models may have biased and inconsistent parameter estimates. This article assembled a comprehensive set of bridge and climate data from the National Bridge Inventory (NBI) and the South Carolina Climatology office. Bridge component deterioration models for bridges on state highways in South Carolina were estimated using an ordered probit model with random effects specification to account for the randomness and panel nature of the bridge data. The study results are useful for various bridge management tasks including maintenance programming, budgeting and bridge asset evaluation.

Methodological Comparison of Mapping the Expanded Prostate Cancer Index Composite to EuroQoL-5D-3L Using Cross-Sectional and Longitudinal Data: Secondary Analysis of NRG/RTOG 0415.

To compare the predictive ability of mapping algorithms derived using cross-sectional and longitudinal data. This methodological assessment used data from a randomized controlled noninferiority trial of patients with low-risk prostate cancer, conducted by NRG Oncology (ClinicalTrials.gov identifier: NCT00331773), which examined the efficacy of conventional schedule versus hypofractionated radiation therapy (three-dimensional conformal external beam radiation therapy/IMRT). Health-related quality-of-life data were collected using the Expanded Prostate Cancer Index Composite (EPIC), and health utilities were obtained using EuroQOL-5D-3L (EQ-5D) at baseline and 6, 12, 24, and 60 months postintervention. Mapping algorithms were estimated using ordinary least squares regression models through five-fold cross-validation in baseline cross-sectional data and combined longitudinal data from all assessment periods; random effects specifications were also estimated in longitudinal data. Predictive performance was compared using root mean square error. Longitudinal predictive ability of models obtained using baseline data was examined using mean absolute differences in the reported and predicted utilities. A total of 267 (and 199) patients in the estimation sample had complete EQ-5D and EPIC domain (and subdomain) data at baseline and at all subsequent assessments. Ordinary least squares models using combined data showed better predictive ability (lowest root mean square error) in the validation phase for algorithms with EPIC domain/subdomain data alone, whereas models using baseline data outperformed other specifications in the validation phase when patient covariates were also modeled. The mean absolute differences were lower for models using EPIC subdomain data compared with EPIC domain data and generally decreased as the time of assessment increased. Overall, mapping algorithms obtained using baseline cross-sectional data showed the best predictive performance. Furthermore, these models demonstrated satisfactory longitudinal predictive ability.

Does earnings quality impact firms’ performance? The case of Portuguese SMEs from the mold sector

PurposeThis study aims to study the impact of earnings quality on firms’ financial performance.Design/methodology/approachAn unbalanced panel data of 237 small- and medium-sized Portuguese companies from the mold industry, using 2010–2018 yearly data was analyzed. While most studies focus only on earnings management when assessing earnings quality, in this study six proxies for earnings quality are used, namely, accruals quality (a proxy for earnings management), earnings persistence, earnings predictability, earnings smoothness, earnings timeliness and earnings conservatism. Moreover, two proxies of financial performance are considered, the return on assets and the economic value added. An econometric model was estimated using either a fixed-effects or a random-effects specification to account for the individual firm-specific effects and ensure heteroscedasticity corrected estimates.FindingsThe results show that managers must be concerned with the quality of reported earnings, as it can affect positively firms’ financial performance, especially regarding accruals quality. Persistence, predictability, smoothness, timeliness and conservatism are shown not to exert significant influence on financial performance in the sample.Research limitations/implicationsThis work contributes not only as a literature review on these thematic but also to firms’ managers and stakeholders, who have information that helps them select strategies that guarantee earnings quality and improve firms’ financial performance.Originality/valueThis study proposed an econometric model that studies the relationship between earnings quality (using several proxies for it) and financial performance that can be applied to all companies.

The association between bed occupancy rates and hospital quality in the English National Health Service

We study whether hospitals that exhibit systematically higher bed occupancy rates are associated with lower quality in England over 2010/11–2017/18. We develop an economic conceptual framework to guide our empirical analysis and run regressions to inform possible policy interventions. First, we run a pooled OLS regression to test if high bed occupancy is associated with, and therefore acts as a signal of, lower quality, which could trigger additional regulation. Second, we test whether this association is explained by exogenous demand–supply factors such as potential demand, and unavoidable costs. Third, we include determinants of bed occupancy (beds, length of stay, and volume) that might be associated with quality directly, rather than indirectly through bed occupancy. Last, we use a within-between random-effects specification to decompose these associations into those due to variations in characteristics between hospitals and variations within hospitals. We find that bed occupancy rates are positively associated with overall and surgical mortality, negatively associated with patient-reported health gains, but not associated with other indicators. These results are robust to controlling for demand–supply shifters, beds, and volume. The associations reduce by 12%-25% after controlling for length of stay in most cases and are explained by variations in bed occupancy between hospitals.

Physician Gender, Patient Risk, and Web-Based Reviews: Longitudinal Study of the Relationship Between Physicians' Gender and Their Web-Based Reviews.

BackgroundWeb-based reviews of physicians have become exceedingly popular among health care consumers since the early 2010s. A factor that can potentially influence these reviews is the gender of the physician, because the physician’s gender has been found to influence patient-physician communication. Our study is among the first to conduct a rigorous longitudinal analysis to study the effects of the gender of physicians on their reviews, after accounting for several important clinical factors, including patient risk, physician specialty, and temporal factors, using time fixed effects. In addition, this study is among the first to study the possible gender bias in web-based reviews using statewide data from Alabama, a predominantly rural state with high Medicaid and Medicare use.ObjectiveThis study conducts a longitudinal empirical investigation of the relationship between physician gender and their web-based reviews using data across the state of Alabama, after accounting for patient risk and temporal effects.MethodsWe created a unique data set by combining data from web-based physician reviews from the popular physician review website, RateMDs, and clinical data from the Center for Medicare and Medicaid Services for the state of Alabama. We used longitudinal econometric specifications to conduct an econometric analysis, while controlling for several important clinical and review characteristics across four rating dimensions (helpfulness, knowledge, staff, and punctuality). The overall rating and these four rating dimensions from RateMDs were used as the dependent variables, and physician gender was the key explanatory variable in our panel regression models.ResultsThe panel used to conduct the main econometric analysis included 1093 physicians. After controlling for several clinical and review factors, the physician random effects specifications showed that male physicians receive better web-based ratings than female physicians. Coefficients and corresponding SEs and P values of the binary variable GenderFemale (1 for female physicians and 0 otherwise) with different rating variables as outcomes were as follows: OverallRating (coefficient –0.194, SE 0.060; P=.001), HelpfulnessRating (coefficient –0.221, SE 0.069; P=.001), KnowledgeRating (coefficient –0.230, SE 0.065; P<.001), StaffRating (coefficient –0.123, SE 0.062; P=.049), and PunctualityRating (coefficient –0.200, SE 0.067; P=.003). The negative coefficients indicate a bias toward male physicians versus female physicians for aforementioned rating variables.ConclusionsThis study found that female physicians receive lower web-based ratings than male physicians even after accounting for several clinical characteristics associated with the physicians and temporal effects. Although the magnitude of the coefficients of GenderFemale was relatively small, they were statistically significant. This study provides support to the findings on gender bias in the existing health care literature. We contribute to the existing literature by conducting a study using data across the state of Alabama and using a longitudinal econometric analysis, along with incorporating important clinical and review controls associated with the physicians.

Bayesian Meta-Analysis to Validate Correlate of Protection for High Vaccine Efficacy Clinical Trials

In clinical trials, a correlate (surrogate) of protection (CoP) endpoint must be properly validated through rigorous sound methods before it may be approved for use. The validation of surrogate in the context of high vaccine efficacy trials, however, poses great challenge due to sparse data; and conventional methods for statistical validation of surrogate are no longer adequate. Although idea of surrogacy was developed in the context of a single trial, the meta-analytic approach, which allows both individual and trial level surrogacy, has become well accepted. However, the meta-analytic joint bivariate full models suffer computational issues. To ease the challenge, aggregate data may be used but it leads to loss of information. In this manuscript the direct application of individual level (instead of aggregate) data in a Bayesian Hierarchical Modelling framework was proposed. The proposed method uses reduced bivariate models with trial specific random effects of treatment on the endpoints and no correlated residuals. Simulated data consist several scenarios, each of which has 5000 participants data, 50 subgroups (used as trials) characterised by size of 100 participants per trial randomised in the ratio 1:1 to vaccinated and unvaccinated treatment groups. The meta-analysis showed improved quality of the CoP compared to literature based on aggregate data. There were no computational issues with the proposed hierarchical model.&#x0D; 2.11.0.0

Bipartite exponential random graph models with nodal random effects

We examine the inclusion of specific nodal random effects for first- and second-mode nodes towards an ERGM for bipartite networks. The inclusion of such node-specific random effects in the ERGM accounts for unobserved heterogeneity in the bipartite network and ensures stable estimation results, especially for large-scale bipartite networks. Moreover, The predicted nodal random effects deliver reasonable interpretation to understand the network behavior. The estimation is carried out by an iterative estimation technique, iterating between pseudolikelihood estimation for the nodal random effects and maximum likelihood estimation for the network parameters.

Application and Comparison of Methods for Analysing Correlated Interval-censored Data from Sexual Partnerships

In epidemiological studies where subjects are seen periodically on follow-up visits, interval-censored data occur naturally. The exact time the change of state (such as HIV seroconversion) occurs is not known exactly, only that it occurred sometime within a specific time interval. This paper considers estimation of parameters when HIV infection times are intervalcensored and correlated. It is assumed that each sexual partnership has a specific unobservable random effect that induces association between infection times. Parameters are estimated using the expectation-maximization algorithm and the Gibbs sampler. The results from the two methods are compared. Both methods yield fixed effects and baseline hazard estimates that are comparable. However, standard errors and frailty variance estimates are underestimated in the expectation-maximization algorithm compared to those from the Gibbs sampler. The Gibbs sampler is considered a plausible alternative to the expectation-maximization algorithm.

Measuring the Attenuation in a Subject-specific Random Effect with Paired Data

This paper is motivated by an investigation into the growth of pigs, which studied among other things the effect of short–term feed with drawal on live weight. This treatment was thought to reduce the variability in the weights of the pigs. We represent this reduction as an attenuation in an animal–specific random effect. Given data on each pig before and after treatment, we consider the problems of testing for a treatment effect and measuring the strength of the effect, if significant. These problems are related to those of testing the homogeneity of correlated variances, and re gression with errors in variables. We compare three different estimates of the attenuation factor using data on the live weights of pigs, and by simulation.

Expenditure Per Capita Model with Spatial Small Area Estimation

Indonesia is one of many countries around the world that attempt to suffer from high poverty rates. Since, poverty information in a certain area is a point of interest to researchers and policy makers. One problem faced is for the development program to be carried out more effectively and efficiently, it is necessary to have data availability up to the micro-scale. The technique used to reach the goal is Small Area Estimation (SAE). Fay-herriot (FH) model is one method on Small Area Estimation. Since, the SAE techniques require “borrow strength” across neighbor areas so thus Fay-Herriot model approach was developed by integrating spatial information into the model. This method known as Spatial Fay-Herriot Model (SFH) or Spatial Empirical Best Linear Unbiased Prediction (SEBLUP). This study aims to compare MSE of direct estimation, FH, and SFH Model to see which method gives the best result in estimating expenditure. The MSE value of the estimated SFH is smaller than direct estimation and FH, but it does not significant. It means adding spatial information in the small area estimation model does not give a better prediction than the simple small area estimation which is takes account the area as a specific random effect.

Exploring efficient linear mixed models to detect quantitative trait locus-by-environment interactions.

Genotype-by-environment (G × E) interactions are important for understanding genotype–phenotype relationships. To date, various statistical models have been proposed to account for G × E effects, especially in genomic selection (GS) studies. Generally, GS does not focus on the detection of each quantitative trait locus (QTL), while the genome-wide association study (GWAS) was designed for QTL detection. G × E modeling methods in GS can be included as covariates in GWAS using unified linear mixed models (LMMs). However, the efficacy of G × E modeling methods in GS studies has not been evaluated for GWAS. In this study, we performed a comprehensive comparison of LMMs that integrate the G × E modeling methods to detect both QTL and QTL-by-environment (Q × E) interaction effects. Model efficacy was evaluated using simulation experiments. For the fixed effect terms representing Q × E effects, simultaneous scoring of specific and nonspecific environmental effects was recommended because of the higher recall and improved genomic inflation factor value. For random effects, it was necessary to account for both G × E and genotype-by-trial (G × T) effects to control genomic inflation factor value. Thus, the recommended LMM includes fixed QTL effect terms that simultaneously score specific and nonspecific environmental effects and random effects accounting for both G × E and G × T. The LMM was applied to real tomato phenotype data obtained from two different cropping seasons. We detected not only QTLs with persistent effects across the cropping seasons but also QTLs with Q × E effects. The optimal LMM identified in this study successfully detected more QTLs with Q × E effects.

Does highway project bundling policy affect bidding competition? Insights from a mixed ordinal logistic model

Abstract Highway transportation project bundling can significantly reduce project delivery cost but could cause undesirable consequences such as reduced market competition. As such, bundling policy needs to be guided by a careful assessment of the potential impacts of bundling on bidding competition. This paper analyses the effect of project bundling policies on market competition using random-effects mixed ordinal logistic modelling, a probabilistic approach. The paper also investigates the variations of such effects across the different project types. The bundling policy factors that were found to be significantly impact bidding competition were the bundle size, bundling combination, and spatial proximity. The results suggest that bundled projects are generally associated with lower bidding competition compared to standalone projects, for all work categories except traffic projects. Further, where the bundled projects have proximal location, the bidding competition is generally higher. It was also determined that the different highway project types exhibit significant differences regarding the effect of bundling on competition: bundling policy (vis-a-vis standalone policy) generally leads to less intense competition for bridge, road, small-structures, miscellaneous, and utility projects, and more intense competition for traffic projects. In addition, it was found that there is a significant improvement on the model performance by including a random effects specification which diagnosed and addressed the unobserved heterogeneity that exists across the contract-letting years, for all the project types. The paper’s models can help highway agencies not only estimate the expected number of bidders for a given project under different bundling policies but also assess the effects of the bundling policy-related factors on bidding competition.

Hierarchical random effects model for insurance pricing of vehicles belonging to a fleet

We propose a hierarchical random effect model for the posterior insurance ratemaking of vehicles belonging to a fleet by allowing random effects for fleet, vehicle, and time. The model is an alternative to the gamma-Dirichlet model of Angers et al (2018), which does not allow for a closed form posterior ratemaking formula. Our theoretical extension derives a simple and tractable closed form ratemaking formula based on a hierarchical random effects specification. We estimate the corresponding econometric model and compute insurance premiums in relation to the past experience of both the vehicle and the fleet. Our econometric model can also be applied to any other dynamic count modeling application with random individual, time, and common effects, such as labor contracting, chirurgical accidents, or any other random event implying principals and many agents.

NONLINEAR PANEL DATA MODELS WITH DISTRIBUTION-FREE CORRELATED RANDOM EFFECTS

Under a Mundlak-type correlated random effect (CRE) specification, we first show that the average likelihood of a parametric nonlinear panel data model is the convolution of the conditional distribution of the model and the distribution of the unobserved heterogeneity. Hence, the distribution of the unobserved heterogeneity can be recovered by means of a Fourier transformation without imposing a distributional assumption on the CRE specification. We subsequently construct a semiparametric family of average likelihood functions of observables by combining the conditional distribution of the model and the recovered distribution of the unobserved heterogeneity, and show that the parameters in the nonlinear panel data model and in the CRE specification are identifiable. Based on the identification result, we propose a sieve maximum likelihood estimator. Compared with the conventional parametric CRE approaches, the advantage of our method is that it is not subject to misspecification on the distribution of the CRE. Furthermore, we show that the average partial effects are identifiable and extend our results to dynamic nonlinear panel data models.

Semi-parametric survival analysis via Dirichlet process mixtures of the First Hitting Time model.

Time-to-event data often violate the proportional hazards assumption inherent in the popular Cox regression model. Such violations are especially common in the sphere of biological and medical data where latent heterogeneity due to unmeasured covariates or time varying effects are common. A variety of parametric survival models have been proposed in the literature which make more appropriate assumptions on the hazard function, at least for certain applications. One such model is derived from the First Hitting Time (FHT) paradigm which assumes that a subject's event time is determined by a latent stochastic process reaching a threshold value. Several random effects specifications of the FHT model have also been proposed which allow for better modeling of data with unmeasured covariates. While often appropriate, these methods often display limited flexibility due to their inability to model a wide range of heterogeneities. To address this issue, we propose a Bayesian model which loosens assumptions on the mixing distribution inherent in the random effects FHT models currently in use. We demonstrate via simulation study that the proposed model greatly improves both survival and parameter estimation in the presence of latent heterogeneity. We also apply the proposed methodology to data from a toxicology/carcinogenicity study which exhibits nonproportional hazards and contrast the results with both the Cox model and two popular FHT models.