Correlated Error Terms Research Articles

Factors affecting farmers' choice to adopt risk management strategies: The application of multivariate and multinomial probit models

This study investigates the factors that impact farmers' adoption of risk management strategies (RMS) in Pakistan during times of uncertainty. The study examines farmers' RMS adoption choices using both multinomial probit (MNP) and multivariate probit (MVP). Data were collected from 382 sampled farmers from four districts in Khyber-Pakhtunkhwa (KP) province of Pakistan via a multistage sampling technique. This study utilizes the MNP model, considering the assumption of Independence of Irrelevant Alternatives (IIA) and incorporating correlated error terms. The objective is to understand farmers' behavior in risky situations and determine if there is heterogeneity. Results are compared with the MVP model to assess robustness and gain insights into farmers' decision-making processes.The research findings reveal that our results are robust, and farmers behave homogeneously in different settings of the provision of risk management strategies. Farmers adopt RMS individually or in combination to mitigate the adverse effects of natural calamities on their livelihood. The risk-averse farmers, who perceive weather-related risks as a threat, access credits and information, and have farms close to a river are more likely to adopt RMS, irrespective of the format of the strategies available. Moreover, the predicted probabilities and correlation of the RMS and RM categories further strengthened our model estimation. These findings provide insights into the behavior of farmers in adopting RMS which are helpful for policymakers and stakeholders in developing strategies to mitigate the impacts of natural calamities on farmers.

A bootstrap-based approach to combine individual-based forest growth models and remotely sensed data

Abstract Combining forest growth models with remotely sensed data is possible under a generalized hierarchical model-based (GHMB) inferential framework. This implies the existence of two submodels: the growth model itself ($\mathcal{M}_{1}$) and a second submodel that links the growth predictions to some remotely sensed variables ($\mathcal{M}_{2}$). Analytical GHMB estimators are available to fit submodel $\mathcal{M}_{2}$ and account for the uncertainty stemming from submodel $\mathcal{M}_{1}$, i.e. the growth model. However, when the growth model is individual based, it is usually too complex to be differentiated with respect to its parameters. As a result, the analytical GHMB estimators cannot be used. In this study, we developed a bootstrap approach for the GHMB inferential framework in order to combine individual-based forest growth models with remotely sensed data. Through simulation studies, we showed that the bootstrap estimators were nearly unbiased when both submodels were linear. The estimator of the parameter estimates remained nearly unbiased when submodel $\mathcal{M}_{1}$ became complex, i.e. non-differentiable, and submodel $\mathcal{M}_{2}$ was nonlinear with heterogeneous variances and correlated error terms. The variance estimator showed some biases but these were relatively small. We further demonstrated through a real-world case study that the predictions of a complex individual-based model could be linked to a Landsat-8 near-infrared spectral band in the boreal forest zone of Quebec, Canada.

Psychometric properties of the Perceived Stress Scale-4 (PSS-4) in a Colombian sample: Onefactor, two-factor, or method effects?

Introduction: The Perceived Stress Scale-4 (PSS-4) is an ultra-brief self-report measure to assess psychological stress. This study evaluated the psychometric properties of the PSS-4 in Colombia. Method: A total of 1,911 adult participants with a university education completed the PSS-4. The characteristics of the items and the subscales were explored. The dimensionality was assessed using principal component analysis and confirmatory factor analysis (CFA), incorporating an examination of invariance (configural, metric, scalar, and strict) across socio-demographic characteristics. Construct validity (convergent and discriminant), reliability indices, know-groups, and predictive validity were also computed. Results: CFA showed that the one-factor model (psychological stress) with method effects (correlated error terms on the negative-phrased items) was the most appropriate according to fit indices and parsimony considerations, being invariant across gender and age, but not across income level and work status. Internal consistency was adequate for the PSS-4. Significant positive correlations of the PSS-4 were identified with depression and anxiety, as well as significant negative correlations were identified with post-traumatic growth and resilience. The PSS-4 showed adequate capacity to predict potential depressive and anxiety symptoms, as well as protective factors such as resilience and post-traumatic growth. Higher scores on the PSS-4 were observed among young people, as well as among people with lower incomes and those who are unemployed. Conclusions: These findings suggest that the PSS-4 can be a reliable and valid tool for assessing psychological stress in Colombians with a university education.

Psychometric validation of the Hospital Anxiety and Depression Scale (HADS) in community-dwelling older adults

ObjectivesThe Hospital Anxiety and Depression Scale (HADS) is commonly used to measure anxiety and depression, but the number of studies validating psychometric properties in older adults are limited. To our knowledge, no previous studies have utilized confirmative factor analyses in community-dwelling older adults, regardless of health conditions. Thus, this study aimed to examine the psychometric properties of HADS in older adults 70 + living at home in a large Norwegian city.MethodsIn total, 1190 inhabitants ≥ 70 (range 70 – 96) years completed the HADS inventory in the population-based Trøndelag Health Study (HUNT), termed “HUNT4 70 + ” in Trondheim, Norway. Confirmatory factor analyses were performed to test the dimensionality, reliability, and construct validity.ResultsThe original two-factor-solution (Model-1) revealed only partly a good fit to the present data; however, including a cross-loading for item 6D (“I feel cheerful”) along with a correlated error term between item 2D (“I still enjoy the things I used to enjoy”) and 12D (“I look forward with enjoyment to things”) improved the fit substantially. Good to acceptable measurement reliability was demonstrated, and the construct validity was acceptable.ConclusionsThe HADS involves some items that are not reliable and valid indicators for the depression construct in this population, especially item 6 is problematic. To improve the reliability and validity of the Norwegian version of HADS, we recommend that essential aspects of depression in older adults should be included.

Forecasting emergency department waiting time using a state space representation.

The provision of waiting time information in emergency departments (ED) has become an increasingly popular practice due to its positive impact on patient experience and ED demand management. However, little scientific attention has been given to the quality and quantity of waiting time information presented to patients. To improve both aspects, we propose a set of state space models with flexible error structures to forecast ED waiting time for low acuity patients. Our approach utilizes a Bayesian framework to generate uncertainties associated with the forecasts. We find that the state-space models with flexible error structures significantly improve forecast accuracy of ED waiting time compared to the benchmark, which is the rolling average model. Specifically, incorporating time-varying and correlated error terms reduces the root mean squared errors of the benchmark by 10%. Furthermore, treating zero-recorded waiting times as unobserved values improves forecast performance. Our proposed model has the ability to provide patient-centric waiting time information. By offering more accurate and informative waiting time information, our model can help patients make better informed decisions and ultimately enhance their ED experience.

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

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

Does the Five Facet Mindfulness Questionnaire measure the same constructs before and after ACT-based treatment? Examination of the longitudinal measurement invariance of the 24-item FFMQ-SF

Violations of longitudinal measurement invariance (LMI) suggest “response shifts” in individuals’ understanding and appraisal of items over time and result in misinterpretation of observed changes in outcome measures. These concerns are crucial to the validity of outcome measures and are particularly relevant in examining the effectiveness of mindfulness and other process-based interventions. In the current report from the Rhode Island Methods to Improve Diagnostic Assessment and Services (MIDAS) project, we examine the LMI of the 24-item short form of the Five Factor Mindfulness Questionnaire from intake to discharge in a large, diagnostically heterogeneous sample of psychiatric patients (n = 2928) participating in an acceptance and commitment therapy-based partial hospitalization program. The original five-factor structure required correlated error terms for items with similar wording (Nonjudging of Experience and Describe scales) to achieve acceptable fit. Following these modifications, we found evidence of configural and weak invariance for all scales. We found evidence for strong invariance for Nonjudging of Experience, Nonreactivity to Inner Experiences, Describe, and Observing, but not for Acting with Awareness. Strict invariance was not supported. Results highlight the challenges of measuring constructs involving overt metacognitive processes via self-report. Implications for scale development and clinical practice are discussed.

The relationship between emotion and cognition in cognitively healthy cohort: Using NIH toolbox emotion battery

AbstractBackgroundThe NIH Toolbox was designed to provide brief, reliable, and valid measures of neurological and behavioral functions. Emotion assessment using the NIH Toolbox Emotion Battery has recently received significant attention given emerging evidence that emotional health is predictive of physical health and overall well‐being. The purpose of this study is to examine whether the modified emotion toolbox model is associated with cognition in sample of cognitively healthy adults.MethodParticipants were 529 enrollees in the Wisconsin ADRC who had normal cognition. The majority of participants were female (65.8%) and White (78.3%). Average age was 63.69 (SD = 8.37). NIH Toolbox Emotion Battery factors and a measure of executive function, Trail Making Test (TMT) Parts A & B were examined using SEM. To evaluate the overall goodness‐of‐fit, we used the following criteria: χ2 nonsignificant, χ2/df between1‐3, comparative fit index (CFI)>0.95, Tucker–Lewis index (TLI)>0.95, standardized root mean square residual (SRMR)<0.05, and the root mean square error of approximation (RMSEA)<0.08.ResultThe initial SEM results indicated a non‐acceptable fit for the data: χ2(51,N = 529) = 275.99, p<.01 (should be not significant), χ2/df = 5.41 (<3), CFI = 0.92(>.95), TLI = 0.90 (>.95), SRMR = 0.05, and RMSEA = 0.09 (<.08), 90% confidence interval (CI) = [0.08, 0.10] based on previously reported goodness‐of‐fit criteria. Based on modification indices, conceptually and statistically meaningful correlated error terms were added to the model (e,g., proximity between emotional support and friendship; emotional support and instrumental support; stress and sadness). After correlating some error terms based on theoretical background and modification indices, the modified SEM results indicated an acceptable fit for the data: χ2/df = 2.52 (<3), CFI = 0.98, TLI = 0.97, SRMR = 0.03 and RMSEA = 0.05, 90% confidence interval (CI) = [0.04,0.06].ConclusionOur modified model using the NIH Toolbox Emotion Battery had an acceptable fit for the data in a cognitively healthy cohort. Although we did not include all emotion toolbox variables, our modified model still had a great fit to the data. We specifically found that positive emotions including social satisfaction measures were correlated with cognition. We did not find any correlation between negative affect and cognition, which could be due to limited sample size. Measuring and identifying positive and negative emotions are important in clinical and research practice.

A comparison of artificial neural networks and regression modeling techniques for predicting dominant heights of Oriental spruce in a mixed stand

Aim of study: This paper introduces comparative evaluations of artificial neural network models and regression modeling techniques based on some fitting statistics and desirable characteristics for predicting dominant height.
 Area of study: The data of this study were obtained from Oriental spruce (Picea orientalis L.) felled trees in even-aged and mixed Oriental spruce and Scotch pine (Pinus sylvestris L.) stands in the northeast of Türkiye.
 Material and methods: A total of 873 height-age pairs were obtained from Oriental spruce trees in a mixed forest stand. Nonlinear mixed-effects models (NLMEs), autoregressive models (ARM), dummy variable method (DVM), and artificial neural networks (ANNs) were compared to predict dominant height growth.
 Main results: The best predictive model was NLME with a single random parameter (root mean square error, RMSE: 0.68 m). The results showed that NLMEs outperformed ARM (RMSE: 1.09 m), DVM in conjunction with ARM (RMSE: 1.09 m), and ANNs (RMSE: from 1.11 to 2.40 m) in the majority of the cases. Whereas considering variations among observations by random parameter(s) significantly improved predictions of dominant height, considering correlated error terms by autoregressive correlation parameter(s) enhanced slightly the predictions. ANNs generally underperformed compared to NLMEs, ARM, and DVM with ARM.
 Research highlights: All regression techniques fulfilled the desirable characteristics such as sigmoidal pattern, polymorphism, multiple asymptotes, base-age invariance, and inflection point. However, ANNs could not replicate most of these features, excluding the sigmoidal pattern. Accordingly, ANNs seem insufficient to assure biological growth assumptions regarding dominant height growth.

Cross-section asymptotic for random-effects panel data models with autoregressive errors

A useful strategy to examine serial correlation in panel regression models is to adopt serially correlated error terms. The random effect is also present to address the individual-specific heterogeneity. Familiar transformations can restructure the AR(1) process into a serially uncorrelated dynamic model. It may result in unfitting estimation results once the statistical inference is carried on the conditional likelihood. We discuss in the article the consequence of this action. In particular, in the case of large cross-section units and short time-sequences, we provide analytical expressions for the asymptotic bias of maximum likelihood estimates when the analyst naïvely neglects the initial conditions problem. Under several conditions, we compute the size of biases and measure the mean squared errors to illustrate this violation. We also construct a joint regression model of the initial and subsequent responses to handle the initial conditions. Then, we conduct simulation studies to determine the asymptotic behavior.

Psychometric properties of the School Kindness Scale in Hong Kong, mainland China, and the Philippines

AbstractStudies have shown that the School Kindness Scale (SKS) has adequate psychometric properties in different societies such as Canada, Turkey, and the Philippines. However, there is scarce evidence on the psychometric validity of this scale across multiple societies and educational contexts. This study explores the cross‐national invariance of the SKS among high school students in the Philippines, Hong Kong, and mainland China. Results showed that the modified unidimensional model of school kindness with correlated error terms on item number 2 and 3 had the most optimal fit. There was evidence supporting partial invariance of the modified unidimensional model of school kindness across setting and year level, and full invariance across gender. Whereas school kindness also demonstrated positive correlations with perceived academic performance in Hong Kong and mainland China, this construct was linked to higher emotional and social engagement in math in all contexts.

Estimation of hurst exponent for sequential monitoring of clinical trials with covariate adaptive randomization

BackgroundClassical Brownian motion (BM) has been commonly used in monitoring clinical trials including those with covariate adaptive randomization (CAR). Independent increment property is commonly assumed in the sequential monitoring process of the clinical trials with CAR designs. However, in reality, correlation may exist in the error terms of the underlying model, resulting in dependent increment in the sequential monitoring process. MethodsWe conducted simulations for estimating the Hurst exponent to evaluate the stochastic property in the covariate adaptive randomized clinical trials under two scenarios: 1. CAR designs with independent and identically distributed error terms. 2. CAR designs with correlated error terms. The theoretical properties of covariate adaptive randomized clinical trials with correlated error structure were investigated. A test statistic including the covariance pattern of the error terms was proposed. ConclusionIn our study, the sequential test statistics under CAR procedure is shown to be asymptotically Brownian motion when the error structure is correctly specified. Further, Brownian motion is a special case of fractional Brownian motion when Hurst exponent equals to 0.5. Our simulations are consistent with the theoretical asymptotic results.

Effects of environmental regulation on rural livelihood diversification: Evidence from pastoral China

Previous studies on rural livelihood diversification primarily focused on its determinants and the role of cash transfers. However, little attention has been paid to the effects of environmental regulation schemes on rural livelihood diversification. This study ascertains the impact of China's Grassland Ecological Compensation Policy (GECP) on the income-generating activities and livelihood diversification of its rural recipients. We use a unique national representative survey of 722 herder households in pastoral China. A seemingly unrelated Tobit model is estimated to simultaneously account for the censored nature of the livelihood diversification of recipients and the correlation of error terms of the different livelihood activities. A latent class cluster analysis is performed to split samples into four clusters according to the characteristics of their livelihood strategies. We find a U-shaped relationship between livelihood diversification and GECP subsidies, such that the livelihood diversification index (excluding cash transfers) decreases or increases when the GECP subsidy is below or above 89,500 yuan, respectively. The present compensation levels are far below the turning point, implying that the scheme fails to achieve its objective of diversifying herders' livelihoods into off-farm sectors. Further, the herders' livelihood diversification index increased when cash transfers were included but decreased when they were excluded, reflecting the vulnerability of herders' livelihoods and their strong reliance on the compensation. Finally, the sub-group analysis reveals that cash transfers only influence the livelihood diversification of full-time, but not part-time, pastoralist households.

Study of Inflation using Stationary Test with Augmented Dickey Fuller & Phillips-Peron Unit Root Test (Case in Bukittinggi City Inflation for 2014-2019)

This classical regression model is designed to handle the relationship between stationary variables and should not be applied to non-stationary series. A time series data is said to be stationary if the mean, variance, and covariance remain constant over time. The problem associated with non-stationary variables, and often encountered by researchers when dealing with time series data, is spurious regression. A clear indicator of false regression is the low Durbin-Watson statistic but has a higher coefficient of determination (R2). Therefore, before doing modeling or forecasting using time series data, it is very important to do a stationary test. In this study, we use inflation data in the City of Bukittinggi from January 2014 to December 2019 as a case study. The data shows an uptrend and correlated error terms. Empirical results show that inflation data in Bukittinggi City is a stationary series.

A Spatiotemporal Analysis of the COVID-19 Pandemic in North Africa.

Spatial panel‐data models are estimated to identify the factors of the prevalence of the coronavirus outbreak in North Africa. Using daily data on the number of cases collected between March 2020 and December 2021, three types of general models are investigated, and they include spatial spillovers between the neighboring countries of the region. In one model the spatial dependence is accounted for by adding a spatial lag of the dependent variable (SAR model). In an alternative specification, spatially correlated error terms are considered in the model (SEM), and in the third model a spatial lag dependent variable and spatially correlated errors are both added (SAC). To deal with unobservable individual heterogeneity, random and fixed individual effects specification are investigated in each of these models. The results of the maximum likelihood and generalized method of moments' estimations show that the lift of travel restrictions had an important impact on the spike in the numbers of COVID‐19 cases in North Africa and that the effects of endogenous interactions between the countries are strongly significant. It is found that spatial spillovers and a change in the travel policy are the main factors that can explain the mechanism of spread the coronavirus pandemic in North Africa. However, more data on socio‐demographic and behavioral variables and on vaccination rates are needed to better understand what caused the recent surge in the number of infections in the region.

Statistical modeling and monitoring of image data in the presence of temporal and spatial correlations

In this paper, the regression-based approach was developed to monitor image data under two-scale analysis. In the first scale, wavelet transformation was used to extract the main features of geometric profile created from the images. The next scale was to monitor the small-scale components which could be expressed by correlation in error terms. To monitor correlation in error terms, one parametric and one non-parametric methods were developed. Parameters of the parametric model including spatial correlation coefficient and error term variance were estimated using Ordinary Least Squares (OLS) and Generalized Least Squares (GLS) estimators, respectively. In non-parametric method, no assumption was made about the structure of correlation in error terms. To extract useful information about the nature of correlation in error terms, Functional Principal Component Analysis (FPCA) was used. After extracting features for both scales, some appropriate test statistics were computed. Then, monitoring the process was performed by plotting these test statistics on corresponding control charts. Simulation and industrial case studies were also performed to evaluate the proposed method’s performance in detecting different shifts. The results indicated the proper performance of the proposed method in monitoring industrial processes to detect out-of-control conditions and identify the source of variability.

Internalizing problems, externalizing problems, and prosocial behavior - three dimensions of the strengths and difficulties questionnaire (SDQ): A study among South African adolescents.

This study aimed to examine the factor structure of the Strengths and Difficulties Questionnaire Self-Report version (SDQ-S), its psychometric properties and measurement invariance by gender and language spoken at home, among secondary school students in Western Cape, South Africa. A sample of 3,542 adolescents in Grade 8 (Mean age = 13.7 years) completed the SDQ-S in a three-language questionnaire (Afrikaans, English and isiXhosa). The data were collected from 42 secondary schools in Cape Town, South Africa. Confirmatory factor analyses with the WLSMV estimator with adjustment for cluster effects (schools) were applied. The SDQ-S was originally developed to cover five domains: four "difficulty" domains (hyperactivity/inattention, emotional, conduct, and peer) and one "strengths" domain (prosocial behavior). When the five factors were tested on the data for the current study, poor fit was obtained. After excluding four items, a three-factor solution with no cross-loadings and no correlated error terms obtained acceptable fit. The results are consistent with previous studies. Strong measurement invariance across genders and language spoken at home was confirmed. In studies of community samples, the use of the SDQ-S scale as an instrument with a three-factor dimension (internalizing, externalizing and prosocial) may be more appropriate than using the original five-factor model.

WOMBAT v1.0: a fully Bayesian global flux-inversion framework

Abstract. WOMBAT (the WOllongong Methodology for Bayesian Assimilation of Trace-gases) is a fully Bayesian hierarchical statistical framework for flux inversion of trace gases from flask, in situ, and remotely sensed data. WOMBAT extends the conventional Bayesian synthesis framework through the consideration of a correlated error term, the capacity for online bias correction, and the provision of uncertainty quantification on all unknowns that appear in the Bayesian statistical model. We show, in an observing system simulation experiment (OSSE), that these extensions are crucial when the data are indeed biased and have errors that are spatio-temporally correlated. Using the GEOS-Chem atmospheric transport model, we show that WOMBAT is able to obtain posterior means and variances on non-fossil-fuel CO2 fluxes from Orbiting Carbon Observatory-2 (OCO-2) data that are comparable to those from the Model Intercomparison Project (MIP) reported in Crowell et al. (2019). We also find that WOMBAT's predictions of out-of-sample retrievals obtained from the Total Column Carbon Observing Network (TCCON) are, for the most part, more accurate than those made by the MIP participants.

Riders on the Storm: Hurricane Risk and Coastal Insurance and Mitigation Decisions

This paper utilizes cross-sectional, household-level, survey data combined with data on subjective risk perceptions and experimentally derived risk preferences to analyze the decision to insure against hurricane losses. Our sample encompasses 670 individuals in five states of the United States Gulf Coast Region (Texas, Louisiana, Mississippi, Alabama, and Florida). This study represents one of the few papers to examine wind insurance empirically and the only study to examine flood insurance, wind insurance, and mitigation behavior contemporaneously. Because these decisions are closely related, we employ a mixed-process regression, which allows for correlated error terms across a random-effects bivariate probit model (flood/wind insurance) and a Poisson Log-Normal count model (mitigation). Results indicate positive and statistically significant correlations between the error terms of the insurance and mitigation models but no significant correlation between the error terms of the two insurance models, conditioned on the covariates. We find evidence that risk perceptions and other household factors have some influence on storm risk management, but the strongest effects tend to be related to mandatory insurance requirements associated with location in high-hazard areas.

Classical Approach to Zero-Inflated Dynamic Panel Ordered Probit Model with an Application in Drug Abuse

The Zero inflated ordered categorical data with time series structure are often a characteristic of behavioral research attributed to non-participation decision and zero consumption of substance such as drugs among the participants. The existing Semi-parametric zero inflated dynamic panel probit model with selectivity have exhibited biasness and inconsistency in estimators as a result of poor treatment of initial condition and exclusion of selectivity in the unobserved individual effects respectively. The model assumed that the cut points are known to address heaping in the data and therefore cannot be used when the cut points are unknown. In this paper, a Zero inflated dynamic panel ordered probit models have been developed to address the above challenges. Average partial effects that presents the impacts on the specific probabilities per unit change in the covariates are also given. Since the solutions are not of closed form, Maximum likelihood estimation based on Newton Raphson algorithm was used to estimate the parameters of the model. A Monte Carlo study was carried out to investigate some theoretical properties of the estimators in the models. The study found that the Zero inflated dynamic panel ordered probit models with independent and correlated error terms produced consistent estimators. The Zero inflated dynamic panel ordered probit models with independent and correlated error terms had more accurate estimators than the Dynamic panel ordered probit model. The Zero inflated dynamic panel ordered probit model with correlated error terms fitted the National Longitudinal Survey of Youth 1997 better than Zero inflated dynamic panel ordered probit model with independent error terms and Dynamic panel ordered probit model. The Zero inflated dynamic panel ordered probit model with independent error terms fitted the National Longitudinal Survey of Youth 1997 better the Dynamic panel ordered probit model.