Model Specification Research Articles

A model specification test for nonlinear stochastic diffusions with delay

A model specification test for nonlinear stochastic diffusions with delay

Effects of logistics performance on Vietnam’s exports: a quantitative analysis using the PPML method

This study aims to analyze the effects of logistics performance on Vietnam’s exports at both the overall level and in specific aspects. We investigate the linkage between logistics performance and Vietnam’s exports based on a panel of bilateral trade from Vietnam to 158 trading partners from 2007 to 2018. Using the Poisson pseudo-maximum likelihood method, the positive and significant effects of Vietnam’s and importing countries’ logistics performance on exports are identified. Our estimates become highly consistent after controlling for both time-specific effects and time-invariant country-fixed effects and remain robust to changes in the model specification, contributing reliable empirical evidence to the literature. Considering the specific aspects of logistics performance, we find the significant roles of the ability to track and trace consignments, the quality of logistics services, the customs clearance efficiency, and the timeliness of shipments in trade. From Vietnam’s perspective, improvements in tracking and tracing ability and timeliness can enhance exports to other countries, with larger effects than those of trading partners. Based on these findings, this study provides evidence for policymakers to effectively enhance trade with other countries by developing the national logistics system alongside further logistics cooperation with trading partners and with a greater focus on tracking and tracing ability and timeliness.

The association between body mass index and health-related quality of life in the 2017 and 2018 health survey of England data: A cross-sectional observational analysis.

To provide an updated estimate of the association between body mass index (BMI) and health-related quality of life (HRQoL) among the general population in England and to identify population subgroups with the highest potential utility gains from obesity interventions. The sample included 12 158 adults with valid HRQoL and BMI data from the 2017 and 2018 Health Survey for England. Robust standard error linear regression, controlling for demographic and socioeconomic characteristics, lifestyle behaviours and obesity-related comorbidities, was used for the baseline analysis. Robustness checks assessed the impact of (a) estimator selection; (b) model specifications; (c) statistical outliers at high BMI; (d) potential BMI measurement error; and (e) data pooling. The study found a significant association between HRQoL and BMI, which exhibited an inverted U-shaped relationship. The mean HRQoL peaked at 25.7 kg/m2 in men and 22.6 kg/m2 in women and was reduced in the underweight, overweight and obesity BMI ranges. Sensitivity analyses reported similar coefficients, suggesting a robust model specification. Reduced HRQoL beyond optimal BMI underlines the importance of maintaining a normal BMI range for overall health. The rising prevalence of class III obesity is a major public health concern given its disproportionate impact on health, health care utilization and costs. Obesity management is key to preventing the reduction in HRQoL associated with obesity-related comorbidities, and this analysis supports the development of targeted policies and population health initiatives for people with class III obesity.

Leveraging active queries in collaborative robotic mission planning

This paper focuses on the high-level specification and generation of 3D models for operational environments using the idea of active queries as a basis for specifying and generating multi-agent plans for acquiring such models. Assuming an underlying multi-agent system, an operator can specify a request for a particular type of model from a specific region by specifying an active query. This declarative query is then interpreted and executed by collecting already existing data/information in agent systems or, in the active case, by automatically generating high-level mission plans for agents to retrieve and generate parts of the model that do not already exist. The purpose of an active query is to hide the complexity of multi-agent mission plan generation, data transformations, and distributed collection of data/information in underlying multi-agent systems. A description of an active query system, its integration with an existing multi-agent system and validation of the active query system in field robotics experimentation using Unmanned Aerial Vehicles and simulations are provided.

Public Financing of Professional Sports Facilities and Drug Asset Forfeiture

Local governments provide many crucial services from limited budgets, yet often subsidize the profitable, private businesses in professional sports leagues in the United States. Policing represents one important public service. Policing typically constitutes large portions of government budgets and also generates revenue through fines and forfeitures. Existing evidence suggests that large municipal expenditures in other areas can have an ambiguous effect on policing outcomes. This paper addresses the question of whether large public expenditures on sports facilities affect drug asset forfeiture using two-way fixed effects (TWFE) and generalized dynamic model specifications (GDMS). The results are similar across estimation methods, with static TWFE results suggesting a treatment effect of $1,274–5,589 in additional forfeiture per million in subsidies and results from the newer GDMS estimators suggesting $7,703 per million in subsidies. The results imply that, beyond generating no tangible local economic benefits, public subsidization of sports facilities also leads police to make up budget shortfalls by more aggressive policing, which has important implications for racial equity.

Impact of Urbanization on Environmental Eminence: Moderating Role of Renewable Energy

In the era of modernization, the movement of the multitude from rural to urban regions all over the globe is rising swiftly. This movement crafts so many socio-economic prospects for the masses. However, in chorus, it has made severe challenges for the eminence of the environment due to a decrease in forestation and the arrangement of more buildings and plants, causing CO2 emissions. It is unmanageable to edge the endurance of urbanization, and the issue is how we can switch its adversative effects on the environment. This study investigated the moderating role of renewable energy consumption in the urbanization-CO2 nexus. For this study, twenty-three of the most urbanized economies from around the world were chosen from 1997 to 2021. Three econometrics techniques are applied for empirical investigation: fixed effect model, robust least square and panel quantile regression with twelve model specifications. The dependent variable is carbon dioxide (CO2) emissions. The explanatory variables are gross fixed capital formation, patent application, inflation, financial development, industrial growth, urbanization and interaction term of renewable energy and urbanization. To check the robustness of empirical findings, we used four different proxies of (CO2) emissions and three different proxies of urbanization. In our empirical findings, patent application, inflation and industrial growth are positively and significantly associated with all proxies of CO2 emissions. While financial development is inversely and significantly allied with CO2 emissions. The impact of all proxies of urbanization is positive and significant on CO2 production. But the moderating effect of renewable energy on environmental depredation is inverse and significant. It suggests using clean and renewable energy and developing the financial sector to improve the eminence of the environment. Our research aligns with the sustainable development goals and the corporate social responsibility stream, making some valuable contributions to the body of previously established research.

A novel approach to evaluate data integrity: evidence from COVID-19 in China

The COVID-19 pandemic has generated an unprecedented amount of epidemiological data. Yet, concerns regarding the validity and reliability of the information reported by health surveillance systems have emerged worldwide. In this paper, we develop a novel approach to evaluating data integrity by combining the Newcomb-Benford Law with outlier methods. We demonstrate the advantages of our framework using a case study from China. To ensure more robust findings, we employ multiple diagnostic procedures, including three conformity estimates, four goodness-of-fit tests, and two distance measures (Cook and Mahalanobis). To promote transparency, we have made all computational scripts publicly available. Our findings indicates a significant deviation in the distribution of new deaths from the theoretical expectations of Benford's Law. Importantly, these results remain accurate even when considering alternative model specifications and conducting various statistical tests. Furthermore, the procedures developed here are easily applicable in other areas of knowledge and can be scaled to assess data quality in both the public and private sectors.

A validity test for a multivariate linear measurement error model

A criterion is proposed for testing the hypothesis about the nature of the error variance in the dependent variable in a linear model, which separates correctly and incorrectly specified models. In the former one, only the measurement errors determine the variance (i.e., the dependent variable is correctly explained by the independent ones, up to measurement errors), while the latter model lacks some independent covariates (or has a nonlinear structure). The proposed MEMV (Measurement Error Model Validity) test checks the validity of the model when both dependent and independent covariates are measured with errors. The criterion has an asymptotic character, but numerical simulations outlined approximate boundaries where estimates make sense. A practical example of the test’s implementation is discussed in detail – it shows the test’s ability to detect wrong specifications even in seemingly perfect models. Estimations of the errors due to the omission of the variables in the model are provided in the simulation study. The relation between measurement errors and model specification has not been studied earlier, and the proposed criterion may stimulate future research in this important area.

Family ownership, finance for innovation, and exports. Evidence from European micro and small firms

ABSTRACT This study focuses on the impact of ownership and the finance for innovation on the exports of micro and small family-owned firms (FFs) compared with nonfamily-owned firms (non-FFs) for 11 European Union countries from 2014 to 2020. The motivation for the study lies in the fact that when combining ownership and innovation, the net effect on exports is not always predictable. Although family ownership negatively affects exports, the gains from innovation are significantly higher for non-FFs than FFs when studying the extensive margin of exports. This implies that the innovation effect is not sufficient to close the export gap for FFs. Regarding the intensive margin of exports, the magnitude of the innovation effect does not differ by firm-type. Thus, FFs show a low intensive margin just because of their ownership. The results hold for different model specifications and individual countries.

The Effect of Terrorism on Income Inequality

Abstract Conflict research has recently found that increased inequality weakens institutional conditions, opening the door for terrorism. While this research often accounts for endogeneity, or the possibility of a reverse causal relationship, there has yet to be an empirical investigation of the impact of terrorism on income inequality. Using a sample of 139 countries between 1994 and 2018, we show that both domestic and transnational terrorism result in higher levels of income inequality. Our results are consistent for both pre- and post-tax and transfer inequality as well as multiple model specifications. Additionally, we explore multiple potential causal mechanisms that link terrorism and income inequality finding that the effect is, in part, due to terrorism’s negative effects on institutional stability, increased military expenditure, possibly to fund counterterrorism operations, and reduced foreign capital, especially for greenfield investment. Our analysis indicates a positive relationship between terrorism and inequality and the mechanisms that produce the effect, contributing to the terrorism and inequality literatures.

Relative income concerns and smoking behaviour: The role of unobserved heterogeneity.

Status or relative concerns (as in the idiom 'keeping up with the Joneses') can lead to negative feelings such as stress and anxiety. One key question is whether these concerns relate to daily smoking behaviour. The conjecture is that status concerns and the accompanying stress and anxiety might be associated with a higher likelihood of smoking and a higher number of cigarettes smoked, generating a higher instant physical reward and reducing the stress and anxiety. The literature aiming to identify this relationship focuses mostly on a single cross section of individuals, ignoring potential differences in unobserved characteristics of smokers and non-smokers (e.g., genetic factors, personality differences, parental smoking during childhood). This paper investigates the role of unobserved individual characteristics on this relationship, which has not been done in previous studies. Using a long panel data of smoking information in Germany and a variety of panel data model specifications, we show that there is no statistically significant association between relative income concerns and the likelihood of smoking or the number of cigarettes smoked among the overall population. We find a positive and significant relationship only among people who smoked at least one cigarette in the past. A 10% appreciation in the income of comparable other individuals relates to about 3.5 more cigarettes per month among these people. Importantly, failing to allow for the unobserved influences of smoking leads to three times larger estimates than when using models with unobserved factors correlating to the income and smoking behaviour. The results are robust with respect to alternative assumptions and specifications where we use different functional forms of unobserved heterogeneity, definitions of relative concerns, incomes, and reference groups.

Can digital transformation enhance corporate ESG performance? The moderating role of dual environmental regulations

This research investigates the influence of digital transformation on corporate ESG performance, with emphasis on dual environmental regulations’ moderating effects, using data from Chinese companies (2011–2020). The results show that digital transformation significantly boosts ESG performance, a finding that persists across various model specifications. The study highlights the amplifying role of both formal and informal environmental regulations in this dynamic, with informal regulations having a stronger effect. This work fills a gap in literature and suggests that environmental regulations are key in leveraging digital transformation for improved ESG outcomes, providing important policy insights.

Enhanced doubly robust estimation with concave link functions for estimands in clinical trials

ABSTRACT For observational studies or clinical trials not fully randomised, the baseline covariates are often not balanced between the treatment and control groups. In this case, the traditional estimates of treatment effects are biased, and causal inference method is needed to get unbiased estimate. The doubly robust estimator (DRE) is a recent popular development in casual inference. However, the unbiasedness of DRE relies on the correct specification of either propensity score or outcome models, which is hardly guaranteed. To overcome this issue, Yuan et al. [(2021), ‘Enhanced Doubly Robust Procedure for Causal Inference’, Statistics in Bioscience, 13(3), 454–478. https://doi.org/10.1007/s12561-021-09300-y.] proposed an enhanced doubly robust estimator which utilises semiparametric models and nonparametric monotone link functions for both propensity score and outcome models. In this article, we further develop an enhanced doubly robust estimator with concave link functions for both propensity score and outcome models. The asymptotic properties of the enhanced doubly robust estimator are studied. Simulation studies are conducted to evaluate the proposed method. A clinical trial data analysis is used to illustrate the method.

MoST: model specification test by variable selection stability

One of the major challenges in empirical studies is to construct an appropriate statistical model that accounts for the uncertainty of statistical methods and model specifications. An improperly specified model may bring the potential risk for the interpretation of conclusions and subsequent decision-making. In this paper, we introduce a Model Specification Test (MoST) inspired by the concept of model confidence bounds and variable selection deviation. To obtain the p-value of our proposed test, we develop an efficient single-layer bootstrap procedure. Our method can be readily applied to existing variable selection strategies without additional assumptions. Extensive numerical experiments demonstrate the feasibility and interpretability of our approach in various scenarios.

The impact of health on economic growth: A narrative literature review

The nexus between health and economic growth is a dynamic and complex relationship. This article reviews the empirical evidence that has sought to assess the causal impact of health on growth, understood as growth in GDP per capita, and focusing on cross-country and selected single country studies. The review largely provides evidence in favour of a positive effect of population health on economic growth. However, the multitude of the factors at play and the possible bidirectional relationship between health and growth pose a challenge for the quantification of the effect and for the relative importance of the underlying mechanisms.There is notable heterogeneity between studies in the magnitude and, in some cases, even in the sign of the effect. The evidence suggests that the health-growth relationship may depend on three main factors: the sample composition (i.e. a country's demographic stage or GDP per capita); the health dimension considered (e.g. health improvements at different life stages may affect productivity differently); and the model specification (e.g. whether or not initial life expectancy is controlled for in the analysis or the quality of the instrument).These findings advocate for a policy approach that integrates health considerations into economic strategies and emphasizes intersectoral collaboration to maximize the economic returns from improved health outcomes.

Export competition between China and Latin America and the Caribbean in the United States market

AbstractWe investigate export competition between China and Latin America and the Caribbean (LAC) in the United States market between 2002 and 2022. Using a sample of 33 exporters and 10‐digit Harmonised Tariff Schedule (HS) level trade data, we estimate a structural gravity model using an instrumental variable constructed from Chinese exports to eight other industrialised nations. We use a first‐order Taylor‐series expansion à la Baier and Bergstrand (Journal of International Economics, 2009a, 77, 77) to approximate the multilateral price terms pointed out by Anderson and Van Wincoop (American Economic Review, 2003, 93, Article 1). The results show that the impact of Chinese exports on United States imports from LAC is negative and statistically significant across several model specifications, levels of aggregation, and sectors. A percentage increase in imports from China decreased imports from LAC by ca. 0.75%. The displacement effect is ca. 0.32 for manufacturing products, 1.01 for resource‐based products, 1.33 when estimated only for South America, 0.25 for the Caribbean, and not significant for Central America.

Risk-taking in banks: does skin-in-the-game really matter?

The belief that bank capital helps improve stability takes for granted the idea that increases in capital are an incentive to reduce risk-taking because bank owners would have more to lose (skin-in-the-game) if their banks fail. Nevertheless, given the higher cost of capital as compared to debt, it is also possible that increases in capital would lead to higher risk-taking due to the need for banks to boost their returns. In light of these contradictory possibilities, we exploit exogenous variations of capital to empirically investigate the actual effects of capital on risk-taking. Our analyses based on a sample of nearly 1900 US Banking Holding Companies in the 1990–2020 period indicate that increasing capital actually leads to higher risk-taking, which contradicts the skin-in-the-game hypothesis. We show evidence that this relationship could be explained by the consequent increase in funding costs that creates pressure for better returns, which is normally achieved by means of taking higher risk. Our main findings are robust to a number of alternative model and sample specifications.

Model Consistency for Mechanical Design: Bridging Lumped and Distributed Parameter Models With a Priori Guarantees

Abstract Engineering design often involves representation in at least two levels of abstraction: the system-level, represented by lumped parameter models (LPMs), and the geometric-level, represented by distributed parameter models (DPMs). Functional design innovation commonly occurs at the system-level, followed by a geometric-level realization of functional LPM components. However, comparing these two levels in terms of behavioral outcomes can be challenging and time-consuming, leading to delays in design translations between system and mechanical engineers. In this paper, we propose a simulation-free scheme that compares LPMs and spatially discretized DPMs based on their model specifications and behaviors of interest, regardless of modeling languages and numerical methods. We adopt a model order reduction (MOR) technique that a priori guarantees accuracy, stability, and convergence to improve the computational efficiency of large-scale models. Our approach is demonstrated through the model consistency analysis of several mechanical designs, showing its validity, efficiency, and generality. Our method provides a systematic way to compare system-level and geometric-level designs, improving reliability and facilitating design translation.

The impact of misclassifications and outliers on imputation methods

Many imputation methods have been developed over the years and tested mostly under ideal settings. Surprisingly, there is no detailed research on how imputation methods perform when the idealized assumptions about the distribution of data and/or model assumptions are partly not fulfilled. This research looks into the susceptibility of imputation techniques, particularly in relation to outliers, misclassifications, and incorrect model specifications. This is crucial knowledge about how well the methods convince in everyday life because, in reality, conditions are usually not ideal, and model assumptions may not hold. The data may not fit the defined models well. Outliers distort the estimates, and misclassifications reduce the quality of most imputation methods. Several different evaluation measures are discussed, from comparing imputed values with true values or comparing certain statistics, from the performance of classifiers to the variance of estimated parameters. Some well-known imputation methods are compared based on real data and simulations. It turns out that robust conditional imputation methods outperform other methods for real data and simulation settings.

Spatial spillover effects associated with the Sweden Democrats' growing support

AbstractResearch on the growing support of radical right parties cannot ignore the role of spatial, contextual effects and geographical proximity. One important spatial factor is the halo effect, which suggests that a radical right party is more likely to perform well in a homogeneous electoral district surrounded by ethnically diverse areas. Voter behaviour is influenced by certain neighbourhood effects, which can be spatially distant, originating from neighbouring areas. In this way, the presence of groups more inclined to support the radical right in an area can, through interaction and observations of signs or norms, modify an individual's party preference. In this paper, I model the 2018 election results of the Sweden Democrats at the level of the 6004 Swedish electoral districts using the Spatial Durbin Model specification. The spatial model shows a better fit and indicates a spillover effect of the level of education and median wealth in addition to the contextual role of ethnic diversity and its change. Changes in the proportion of non‐Europeans in neighbouring areas have had a particularly strong spillover effect, which is associated with the growing support for the Sweden Democrats.