Correlated Error Terms Research Articles

Spsur: An R Package for Dealing with Spatial Seemingly Unrelated Regression Models

Spatial seemingly unrelated regression (spatial SUR) models are a useful multiequational econometric specification to simultaneously incorporate spatial effects and correlated error terms across equations. The purpose of the spsur R package is to supply a complete set of functions to test for spatial structures in the residual of a SUR model; to estimate the most popular specifications by applying different methods and test for linear restrictions on the parameters. The package also facilitates the estimation of socalled spatial impacts, conveniently adapted to a SUR framework. The package includes functions to simulate datasets with the features decided by the user, which may be useful in teaching activities or in more general research projects. The article concludes with a real data application showing the potential that spsur has to examine the relation of individual mobility over geographic areas and the incidence of COVID-19 in Spain during the first lockdown.

Spatial Seemingly Unrelated Regression Models for Freight Trip Generation by Vehicle Type: Application to the Chennai Metropolitan Area in India

During the last two decades, there has been substantial interest in developing freight trip generation (FTG) models. Most studies consider only truck trips or convert all freight trips into equivalent truck trips. Freight in several large cities is increasingly being moved by smaller vehicles. This calls for modeling FTG by vehicle type. The present research identifies and compares establishment characteristics affecting FTG by different vehicle types. In this context, spatial correlations among nearby establishments and the error-term correlations between independent models by vehicle type become relevant. Based on the Lagrange-Multiplier (LM) tests, we develop non-spatial seemingly unrelated regression (SUR) models for freight trip production (FTP) and spatial SUR models with a spatial lag in the dependent variable to account for both spatial and error-term correlations for freight trip attraction (FTA). The results show that establishment type and size affect FTG by different vehicle types.

Lack of measurement invariance in a widely used Facebook addiction scale may thwart progress in research on social-network-use disorder: A cross-cultural study

The social-network-use disorder (SNUD) was recently proposed as a potentially useful diagnosis in the International Classification of Diseases 11th Revision category “other specified disorders due to addictive behaviors.” Problematic Facebook use is among the most often investigated types of SNUD. Bergen Facebook Addiction Scale (BFAS) is the most widely used measure of problematic Facebook use. However, little data support the scale's measurement invariance across different cultural groups or age groups. The present study investigated the validity and measurement invariance of the BFAS across different countries, different age groups, and different cohorts. BFAS was administered alongside depression, anxiety, and stress measures in a total sample of 5470 university and high school students from three continents and five countries: Algeria, Bangladesh, Pakistan, Poland, and the United Arab Emirates. The scale showed a lack of measurement invariance across different group comparisons. Patterns of error term correlations, factor loadings, reliability coefficients, relationships with criterion variables, as well as prevalence estimates differed in groups of interest. Lack of measurement invariance and consequently substantially different functioning of the scales across samples limits any cross-cultural comparisons and generalizability of studies' results on problematic Facebook use. Potential causes of poor cross-cultural performance of BFAS and consequences for research on SNUD are discussed.

Spatial-nonparametric regression: an approach for monitoring image data

Statistical process control plays a significant role in manufacturing industries as processes become more advanced and manufactured products more complex. Because of the increasing sensitivity of the processes and the inadequacy of the methods based on human inspection, use of product images in statistical process control has been considered by some researchers. In this paper, a regression-based method is developed to monitor image data under two-scale analysis. In the first scale, wavelet transformation is used to extract the main features of geometric profile created from the images. The next scale is to monitor the small-scale components which could be expressed by correlation in error terms. To monitor correlation in error terms, a parametric method is developed. Parameters of the parametric model including spatial correlation coefficient and error term variance are estimated using Ordinary Least Squares (OLS) and Generalized Least Squares (GLS) estimators, respectively. After extracting features for both scales, some appropriate test statistics are computed. Then, monitoring the process is performed by plotting these test statistics on corresponding control charts. Performance of the proposed method is evaluated in terms of run length measures and the difference between the actual value and the estimated value of change-points. Simulation studies are performed on tile images, and the results show the capability of the proposed method in detecting out-of-control conditions and estimating the change-point. The results also indicate the proper performance of the proposed method in monitoring industrial processes to detect out-of-control conditions and identifying the source of variability.

Estimating measurement equivalence of the 12-item General Health Questionnaire across ethnic groups in the UK.

This study investigates the extent to which the GHQ-12 exhibits configural, metric and scalar invariance across six ethnic groups in Britain and Northern Ireland, using the UK Household Longitudinal Study (N = 35 410). A confirmatory factor analysis was carried out on a white British group in order to establish an adequate measurement model. Secondly, a multi-group confirmatory factor analysis was conducted in order to assess measurement invariance. A sensitivity analysis comparing summated and latent means across groups was carried out. Finally, revised estimates of scale reliability were derived using two different methods. A one-factor model including correlated error terms on the negatively phrased items showed superior fit in all ethnic groups. Tests for equal factor loadings and intercepts also showed adequate fit demonstrating metric and scalar invariance. Latent and summated scale estimates of mean group differences were similar for all groups. Scale reliability using McDonald's ω is lower than when using the more conventional Cronbach's α. Reliability across groups is reasonably consistent. We find that the GHQ-12 does not display obvious bias in regard to ethnic groups in the UK and that valid comparisons across these groups can be made for the purposes of population research. Caution is needed when using as a screening tool for individuals.

A robust estimation method for the linear regression model parameters with correlated error terms and outliers

Independence of error terms in a linear regression model, often not established. So a linear regression model with correlated error terms appears in many applications. According to the earlier studies, this kind of error terms, basically can affect the robustness of the linear regression model analysis. It is also shown that the robustness of the parameters estimators of a linear regression model can stay using the M-estimator. But considering that, it acquires this feature as the result of establishment of its efficiency. Whereas, it has been shown that the minimum Matusita distance estimators, has both features robustness and efficiency at the same time. On the other hand, because the Cochrane and Orcutt adjusted least squares estimators are not affected by the dependence of the error terms, so they are efficient estimators. Here we are using of a non-parametric kernel density estimation method, to give a new method of obtaining the minimum Matusita distance estimators for the linear regression model with correlated error terms in the presence of outliers. Also, simulation and real data study both are done for the introduced estimation method. In each case, the proposed method represents lower biases and mean squared errors than the other two methods.

Searching for Something Positive in Future Negative: The Adolescent and Adult Time Inventory-Time Attitudes Scale.

The Adolescent and Adult Time Inventory-Time Attitudes Scale (AATI-TA) measures emotional engagement with the past, present, and future, and scores have been shown to relate meaningfully to health outcomes. For past, present, and future, five items are used to assess both positive and negative attitudes. Although evidence for the hypothesized six-factor solution has been widely reported, some studies have indicated problems with the Future Negative items. Given that a large and growing literature has emerged on the six-factor AATI-TA, and that AATI-TA scores have shown much better and more consistent fit than other temporal psychology measures, we sought to investigate the future negative factor in detail. Secondary analyses were performed on two datasets. The first was a University convenience sample (N = 410) and the second was an adolescent sample (N = 1,612). Confirmatory factor analyses revealed that the fit for the five Future Negative items was poor. Modification indices suggested that a correlated error term between Items 4 and 10 would result in good fit, and this was indeed the case. Models without Item 4 or Item 10 also yielded acceptable fit. Analyses using all four operationalizations of Future Negative (original scale, without Item 4 or Item 10, or with the correlated error between Items 4 and 10) to predict symptoms of anxiety and depression, and emotional self-efficacy revealed minor differences in the predictive validity coefficients. Potential ways forward, including a correlated error term or the dropping or replacement of Item 10, are discussed.

Modeling the impacts of park access on health outcomes: A utility-based accessibility approach

Recent research has underscored the potential for public green spaces to influence individual and societal health outcomes, but empirical measurements of such influences have yielded mixed results to date, with particular disagreement surrounding how access to parks ought to be defined while controlling for alternate explanations. In this paper, we apply a comprehensive measure of park accessibility drawn from random utility choice theory, which avoids arbitrary assertions of proximity while incorporating potentially numerous amenities and attributes of both the parks and the population. We apply this utility-based accessibility measure to correlate Census tract-level obesity and physical activity rate estimates from the Centers for Disease Control and Prevention’s 500 Cities project with tract-level American Community Survey socioeconomic data in New York City, paired with geographic open space data from New York City. Controlling for the socioeconomic variables and spatially correlated error terms, we show a positive and significant relationship between park access and physical activity rates. The data also suggest a negative relationship between park access and obesity rates, beyond what is expected through physical activity and socioeconomics. In doing so, this research contributes a more comprehensive modeling approach for measuring the impact of park access on health, and may improve our understanding of the role parks and access to them can serve in furthering public health objectives.

Nested covariance determinants and restricted trek separation in Gaussian graphical models

Directed graphical models specify noisy functional relationships among a collection of random variables. In the Gaussian case, each such model corresponds to a semi-algebraic set of positive definite covariance matrices. The set is given via a parametrization, and much work has gone into obtaining an implicit description in terms of polynomial (in)equalities. Implicit descriptions shed light on problems such as parameter identification, model equivalence and constraint-based statistical inference. For models given by directed acyclic graphs, which represent settings where all relevant variables are observed, there is a complete theory: All conditional independence relations can be found via graphical $d$-separation and are sufficient for an implicit description. The situation is far more complicated, however, when some of the variables are hidden (or in other words, unobserved or latent). We consider models associated to mixed graphs that capture the effects of hidden variables through correlated error terms. The notion of trek separation explains when the covariance matrix in such a model has submatrices of low rank and generalizes $d$-separation. However, in many cases, such as the infamous Verma graph, the polynomials defining the graphical model are not determinantal, and hence cannot be explained by $d$-separation or trek-separation. In this paper, we show that these constraints often correspond to the vanishing of nested determinants and can be graphically explained by the (more general) notion of restricted trek separation.

Bootstrap tests for structural breaks when the regressors and the serially correlated error term are unstable

AbstractStructural break problems in linear regression coefficients are often accompanied by the instabilities of the regressors and/or the serially correlated error term such as structural changes in their marginal distributions. In these circumstances, we find that existing tests as well as Hansen (2000)'s fixed regressor bootstrap have serious size distortions, although the latter is designed to overcome the unstable regressors. To tackle this problem, we propose a method that combines the fixed regressor bootstrap and the sieve wild bootstrap. We show that the sieve wild bootstrap asymptotically replicates a broad set of serially correlated unstable error processes. Using that the fixed regressor bootstrap is designed to capture the instability of the regressors, the mixture of the two is then shown to provide correct asymptotic critical values of the existing tests in various feasible nonstandard cases. Monte Carlo experiments show significant improvements in size and power.

Autoregressive Model With Spatial Dependence and Missing Data

We study herein an autoregressive model with spatially correlated error terms and missing data. A logistic regression model with completely observed covariates is used to model the missingness mechanism. An autoregressive model is used to accommodate time series dependence, and a spatial error model is used to capture spatial dependence. To estimate the model, a weighted least squares estimator is developed for the temporal component, and a weighted maximum likelihood estimator is developed for the spatial component. The asymptotic properties for both estimators are investigated. The finite sample performance is assessed through extensive simulation studies. A real data example about Beijing’s PM level data is illustrated.

The Influence of Peers in Worker Misconduct: Evidence from Restaurant Theft

Problem definition: Misconduct, such as theft, is a major problem in operational settings, and staffing decisions can either amplify or mitigate this problem as workers influence their peers’ behavior. Peers are known to influence coworker productivity and likely also affect counterproductive behaviors. Academic/practical relevance: Many studies have shown how mechanisms such as helping, knowledge transfer, teaching, and social pressure generate productivity peer effects in service and other settings. Yet few papers empirically examine these effects in counterproductive behaviors. We argue that while the same mechanisms driving productivity spillovers also generate peer effects in misconduct, an additional effect—strategic peer response—reflects how coworkers, under managerial monitoring, adjust misconduct in response to peers’ daily behavior. An additional contribution of this paper is to identify the effect of peers on operational performance in a firm setting. Methodology: We use transaction and theft data from 83,153 servers at 1,049 restaurants across 46 states in the United States. We employ instrumental variable (IV) models to account for both reflection problems and correlated error terms in same-day peer theft. We use Monte Carlo simulations to present how biases identified by a combination of ordinary least squares (OLS) and IV models suggest that managerial oversight might generate negative correlation in the same-day error terms of peers that reflects strategic peer responses. Results: Our results show that although servers are more likely to steal when working with high-theft peers, they steal less as peers steal more on a given day. We also show that this negative correlation in daily peer theft is higher under an information technology system that increases managerial oversight by reporting likely theft to managers. Importantly, we demonstrate how reflection effects can significantly amplify even small peer-effect coefficients to have large organizational implications. Our parameter estimates indicate that doubling a single worker’s average theft amount will increase total theft in an average restaurant by 76%. Doubling all workers’ theft amounts increases totals by 550%. Finally, we show that the positive peer effect from high-theft coworkers only exists for new workers in their first three to five months on the job, consistent with imprinting mechanisms that include knowledge transfer and norms. Managerial implications: The results show that the costs of employing unethical workers is higher than the direct cost of those workers’ misconduct because their behavior spills over into coworkers’ actions and amplifies through reflection effects. Yet our results also suggest that this contagion can be mitigated by managerial oversight. So long as there is sufficient monitoring of misconduct, workers will strategically limit such behavior in response to peers.

Psychometric Properties of a Brief Metamemory and Metaconcentration Scale in Substance Use Problem

Dysfunctions of metacognitive aspects of metamemory and metaconcentration are common in substance use. No questionnaire tool has been validated among substance users to assess these metacognitive abilities. Therefore, this study investigated the psychometric properties of a brief metamemory and metaconcentration scale (BMMS) in adults with substance use. Using a cross-section design, participants (n = 395, age = 18–45 years, 22.7 ± 2.4 kg/m2) were purposively selected from houses earmarked by simple random sampling in Mizan, south-west Ethiopia. Interviewer-administered BMMS, the severity of dependence on khat (SDS-khat), and a socio-demographics tool were employed. A two-factor model with correlated error terms was found valid based on adequate model fit and measurement invariance across gender groups. No major ceiling or floor effect was found in the BMMS scores. Good internal consistency (Cronbach’s alpha = 0.83) and adequate internal homogeneity (moderate to strong item-total ISI score correlations; r ≥ 0.46) were found. Khat dependent and non-dependents differed significantly across BMMS total score, factor scores and all the item scores except three. The BMMS has adequate psychometric validity in the population with substance use.

Interfuel substitution, hydroelectricity consumption and CO2 emissions mitigation in Malaysia: evidence from a transcendental logarithm (trans-log) cost function framework.

The main objective of this paper is to estimate the interfuel substitution elasticities between hydropower and the fossil fuels of coal and natural gas used in the generation of electricity for Malaysia. Due to the violation of the assumption behind the ordinary least squares (OLS) method on account of the correlated error terms in the system of equations, the econometrics techniques of seemingly unrelated regression (SUR) was adopted to obtain the parameter estimates using dataset that covers the period 1988 to 2016. The main finding is that there exists substantial substitution possibility between hydropower and fossil fuels in the generation of electricity for Malaysia. CO2 emissions mitigation scenarios were also conducted to explore the possible effects of substituting fossil fuels for hydropower to generate electricity. The results show that switching from high carbon-emitting fuels to renewable energy such as hydropower will substantially reduce CO2 emission and assist the country towards achieving the carbon emissions reduction targets. Policy recommendations are offered in the body of the manuscript.

Testing the Predictability of U.S. Housing Price Index Returns Based on an IVX-AR Model

We use ten common macroeconomic variables to test for the predictability of the quarterly growth rate of house price index (HPI) in the United States during 1975:Q1–2018:Q2. We extend the instrumental variable based Wald statistic (IVX-KMS) proposed by Kostakis, Magdalinos, and Stamatogiannis to a new instrumental variable based Wald statistic (IVX-AR) which accounts for serial correlation and heteroscedasticity in the error terms of the linear predictive regression model. Simulation results show that the proposed IVX-AR exhibits excellent size control regardless of the degree of serial correlation in the error terms and the persistency in the predictive variables, while IVX-KMS displays severe size distortions. The empirical results indicate that the percentage of residential fixed investment in GDP is fairly a robust predictor of the growth rate of HPI. However, other macroeconomic variables’ strong predictive ability detected by IVX-KMS is likely to be driven by the highly correlated error terms in the predictive regressions and thus becomes insignificant when the proposed IVX-AR method is implemented. Supplementary materials for this article, including a standardized description of the materials available for reproducing the work, are available as an online supplement.

Psychometric Characteristics of Childbearing Motivation Questionnaire: A Review

Background: Considering the important role of understanding the childbearing motivations in counseling services about childbearing behavior, and need for applying a valid and reliable instrument to assess it in Iranian society, this study was conducted to determine Childbearing motivation. was validated in two versions including Persian and English. The review aimed to comprehensively review the validity and reliability of Childbearing Motivation to provide comprehensive information to health providers. Materials and Methods: We conducted an extensive search on online databases (Medline, EMBASE, Web of Science, Scopus, Cochrane Library, and CINAHL) from inception until September 2019. The following keywords were used: (Childbearing Questionnaire OR CBQ) AND (Factor Analysis OR Exploratory Factor Analysis OR Confirmatory Factor Analysis OR OR Psychometric OR OR Test-retest Reliability OR Correlation Coefficient OR ICC). COSMIN checklist was used to assess the methodological quality of studies. Results: Three studies were included in review. Intra-class correlation coefficient (ICC), Cronbach's alpha coefficient and the test–retest reliability indicated satisfactory reliability of the CBQ. Original version of Childbearing Motivation was developed by Miller (1995). In Persian version, original model did not show the ‘acceptable’ fit. After removing eight factors with low loading and two correlated error terms, modified model was found to have adequate fitness (Comparative Fit Index (CFI=.91), Tucker–Lewis Index (TLI=.90), Root Mean Square Error of Approximation (RMSEA=.04), and Chi-square to degree of freedom x2/df=1.92). Conclusion: The Persian and English version of childbearing motivation has good construct validity and reliability. Future research needs to test six-factor model with 49 total items and modified model with 43 items in various populations.

Cognitive changes in women starting anticholinergic medications for overactive bladder: a prospective study.

To assess cognitive changes in women 12months after starting anticholinergic medications for overactive bladder syndrome (OAB). We present a prospective cohort study assessing changes in cognition in women seen in a referral urogynecology practice. We compared women who started anticholinergic OAB medications with women not on anticholinergic OAB medications. The primary outcome was change over time on the Montreal Cognitive Assessment (MOCA) screening score. At enrollment, women completed a baseline MOCA screening, a Geriatric Depression Screen (GDS), and an assessment of medications to create an anticholinergic burden score (ACB). At 3, 6, 9, and 12months after enrollment women were administered the MOCA, GDS, and a review of their medications and medical problems. Statistical analysis was performed using a linear mixed effects model taking into account correlated error terms given multiple MOCA assessments at various time points per patient. A total of 106 women were enrolled, 60 in the OAB medication group and 46 in the control (non-OAB medication) group. The mean age was 77years, 93% of women were Caucasian, and 98% completed high school, with no difference between groups. Over time there was no difference in change of MOCA score between the OAB and control groups when controlling for age, GDS score, and ACB score (p = 0.78). This association did not change when women with a neurological diagnosis were excluded (n = 6). On average MOCA scores for the OAB group increased by 0.76 over 12months and the control group increased 0.39, with no difference between the groups (p = 0.53). We found no changes in MOCA scores between OAB medication and control groups after controlling for age, depression, and polypharmacy after 12months of follow-up.

The Indirect Influence of Organizational Safety Climate on Football Players' Concussion Reporting Intentions.

Most of the health interventions designed to increase athletes' reporting of potential concussion symptoms have focused only on the individual athlete. Unfortunately, interventions targeted at athletes' knowledge and understanding of concussion risk has not reliably increased their reporting behavior, leading to increased calls for "changing the culture of concussion reporting." To date, no studies have examined the role of organizational safety climate has on concussion-symptom reporting behavior. We hypothesized that players' perception of organizational safety climate would be indirectly related to their concussion symptom reporting intentions, via the influence of safety culture on supportive social norms and self-efficacy, two well-known predictors of concussion reporting intentions. We used structural equation modeling techniques to create robust latent measures of our model variables and then examined the indirect influence of football program safety climate on football players' symptom reporting intentions. Surveys were completed by NCAA Division I football players (N = 223) before and after the 2017 football season. We tested a confirmatory factor analysis and hypothesized latent variable model with preseason data, made small adjustments to our model (adding correlated error terms), and then confirmed using postseason data. We also examined a competing, alternative model. Results support the indirect and influence of perceived safety climate on concussion reporting intentions primarily via the relationship between safety climate and social norms, and to a lesser extent between safety climate and self-efficacy. Discussion focuses on the importance of considering the addition of interventions aimed at systems-level influences to facilitate supportive social norms and athlete self-efficacy regarding concussion symptom reporting.

Direction selection in stochastic directional distance functions

Researchers rely on the distance function to model multiple product production using multiple inputs. A stochastic directional distance function (SDDF) allows for noise in potentially all input and output variables. Yet, when estimated, the direction selected will affect the functional estimates because deviations from the estimated function are minimized in the specified direction. Specifically, the parameters of the parametric SDDF are point identified when the direction is specified; we show that the parameters of the parametric SDDF are set identified when multiple directions are considered. Further, the set of identified parameters can be narrowed via data-driven approaches to restrict the directions considered. We demonstrate a similar narrowing of the identified parameter set for a shape constrained nonparametric method, where the shape constraints impose standard features of a cost function such as monotonicity and convexity.Our Monte Carlo simulation studies reveal significant improvements, as measured by out of sample radial mean squared error, in functional estimates when we use a directional distance function with an appropriately selected direction and the errors are uncorrelated across variables. We show that these benefits increase as the correlation in error terms across variables increase. This correlation is a type of endogeneity that is common in production settings. From our Monte Carlo simulations we conclude that selecting a direction that is approximately orthogonal to the estimated function in the central region of the data gives significantly better estimates relative to the directions commonly used in the literature. For practitioners, our results imply that selecting a direction vector that has non-zero components for all variables that may have measurement error provides a significant improvement in the estimator’s performance. We illustrate these results using cost and production data from samples of approximately 500 US hospitals per year operating in 2007, 2008, and 2009, respectively, and find that the shape constrained nonparametric methods provide a significant increase in flexibility over second order local approximation parametric methods.

Validity of the Center for Epidemiologic Studies Depression Scale (CES-D) in Eritrean refugees living in Ethiopia

BackgroundDepression is among the top mental health problems with a major contribution to the global burden of disease. This study aimed at identifying the latent factor structure and construct validity...