scholarly journals Bias from network misspecification under spatial dependence

2019 ◽  
Author(s):  
Timm Betz ◽  
Scott J Cook ◽  
Florian M Hollenbach
Keyword(s):  
Spatial Analysis ◽  
Spatial Models ◽  
Limited Information ◽  
Network Ties ◽  
Spatial Econometric ◽  
Cross Sectional ◽  
Limit Condition ◽  
Spatial Weights Matrix ◽  
Cross Sectional Dependence ◽  
Variable Bias

The pre-specification of the network is one of the biggest hurdles for applied researchers in undertaking spatial analysis. In this letter, we demonstrate two results. First, we derive bounds for the bias in non-spatial models with omitted spatially-lagged predictors or outcomes. These bias expressions can be obtained without prior knowledge of the network, and are more informative than familiar omitted variable bias formulas. Second, we derive bounds for the bias in spatial econometric models with non-differential error in the specification of the weights matrix. Under these conditions, we demonstrate that an omitted spatial input is the limit condition of including a misspecificed spatial weights matrix. Simulated experiments further demonstrate that spatial models with a misspecified weights matrix weakly dominate non-spatial models. Our results imply that, where cross-sectional dependence is presumed, researchers should pursue spatial analysis even with limited information on network ties.

scholarly journals Bias from Network Misspecification Under Spatial Dependence

2020 ◽  
pp. 1-7
Author(s):  
Timm Betz ◽  
Scott J. Cook ◽  
Florian M. Hollenbach
Keyword(s):  
Spatial Analysis ◽  
Spatial Models ◽  
Limited Information ◽  
Network Ties ◽  
Spatial Econometric ◽  
Cross Sectional ◽  
Limit Condition ◽  
Spatial Weights Matrix ◽  
Cross Sectional Dependence ◽  
Variable Bias

Abstract The prespecification of the network is one of the biggest hurdles for applied researchers in undertaking spatial analysis. In this letter, we demonstrate two results. First, we derive bounds for the bias in nonspatial models with omitted spatially-lagged predictors or outcomes. These bias expressions can be obtained without prior knowledge of the network, and are more informative than familiar omitted variable bias formulas. Second, we derive bounds for the bias in spatial econometric models with nondifferential error in the specification of the weights matrix. Under these conditions, we demonstrate that an omitted spatial input is the limit condition of including a misspecificed spatial weights matrix. Simulated experiments further demonstrate that spatial models with a misspecified weights matrix weakly dominate nonspatial models. Our results imply that, where cross-sectional dependence is presumed, researchers should pursue spatial analysis even with limited information on network ties.

Spatial Models in Econometric Research

2021 ◽  
Author(s):  
Luc Anselin
Keyword(s):  
Spatial Interaction ◽  
Asymptotic Properties ◽  
Spatial Models ◽  
Critical Element ◽  
Cross Sectional ◽  
Spatial Lag ◽  
Explanatory Variables ◽  
Spatial Weights ◽  
Spatial Weights Matrix ◽  
Definition Of

Since the late 1990s, spatial models have become a growing addition to econometric research. They are characterized by attention paid to the location of observations (i.e., ordered spatial locations) and the interaction among them. Specifically, spatial models formally express spatial interaction by including variables observed at other locations into the regression specification. This can take different forms, mostly based on an averaging of values at neighboring locations through a so-called spatially lagged variable, or spatial lag. The spatial lag can be applied to the dependent variable, to explanatory variables, and/or to the error terms. This yields a range of specifications for cross-sectional dependence, as well as for static and dynamic spatial panels. A critical element in the spatially lagged variable is the definition of neighbor relations in a so-called spatial weights matrix. Historically, the spatial weights matrix has been taken to be given and exogenous, but this has evolved into research focused on estimating the weights from the data and on accounting for potential endogeneity in the weights. Due to the uneven spacing of observations and the complex way in which asymptotic properties are obtained, results from time series analysis are not applicable, and specialized laws of large numbers and central limit theorems need to be developed. This requirement has yielded an active body of research into the asymptotics of spatial models.

Beyond Policy Diffusion: Spatial Econometric Models of Public Administration

2018 ◽  
Vol 29 (4) ◽  
pp. 591-608 ◽  
Author(s):  
Scott J Cook ◽  
Seung-Ho An ◽  
Nathan Favero
Keyword(s):  
Public Administration ◽  
Spatial Models ◽  
State Level ◽  
Cross Sectional Study ◽  
Econometric Models ◽  
Education Spending ◽  
Spatial Econometric ◽  
Cross Sectional ◽  
Econometric Methods ◽  
Spatial Econometric Models

Abstract Interdependence in the decision-making or behaviors of various organizations and administrators is often neglected in the study of public administration. Failing to account for such interdependence risks an incomplete understanding of the choices made by these actors and agencies. As such, we show how researchers analyzing cross-sectional or time-series-cross-sectional (TSCS) data can utilize spatial econometric methods to improve inference on existing questions and, more interestingly, engage a new set of theoretical questions. Specifically, we articulate several general mechanisms for spatial dependence that are likely to appear in research on public administration (isomorphism, competition, benchmarking, and common exposure). We then demonstrate how these mechanisms can be tested using spatial econometric models in two applications: first, a cross-sectional study of district-level bilingual education spending and, second, a TSCS analysis on state-level healthcare administration. In our presentation, we also briefly discuss many of the practical challenges confronted in estimating spatial models (e.g., weights specification, model selection, effects calculation) and offer some guidance on each.

scholarly journals Symbolic Analysis Applied to the Specification of Spatial Trends and Spatial Dependence

Entropy ◽  
2020 ◽  
Vol 22 (4) ◽  
pp. 466
Author(s):  
Maryna Makeienko
Keyword(s):  
Spatial Dependence ◽  
Housing Prices ◽  
Spatial Models ◽  
Unit Roots ◽  
Symbolic Analysis ◽  
Spatial Econometric ◽  
Cross Sectional ◽  
Economic Relations ◽  
Modeling Strategy ◽  
Space Dependence

This article provides symbolic analysis tools for specifying spatial econometric models. It firstly considers testing spatial dependence in the presence of potential leading deterministic spatial components (similar to time-series tests for unit roots in the presence of temporal drift and/or time-trend) and secondly considers how to econometrically model spatial economic relations that might contain unobserved spatial structure of unknown form. Hypothesis testing is conducted with a symbolic-entropy based non-parametric statistical procedure, recently proposed by Garcia-Cordoba, Matilla-Garcia, and Ruiz (2019), which does not rely on prior weight matrices assumptions. It is shown that the use of geographically restricted semiparametric spatial models is a promising modeling strategy for cross-sectional datasets that are compatible with some types of spatial dependence. The results state that models that merely incorporate space coordinates might be sufficient to capture space dependence. Hedonic models for Baltimore, Boston, and Toledo housing prices datasets are revisited, studied (with the new proposed procedures), and compared with standard spatial econometric methodologies.

Extent Pursuit for Cross-Sectional Dependence in Large Panels

2019 ◽  
Author(s):  
Jiti Gao ◽  
Guangming Pan ◽  
Yanrong Yang ◽  
Bo Zhang
Keyword(s):  
Cross Sectional ◽  
Large Panels ◽  
Cross Sectional Dependence

Psychological Impacts of COVID-19 During the First Nationwide Lockdown in Vietnam: Web-Based, Cross-Sectional Survey Study (Preprint)

2020 ◽  
Author(s):  
Khanh Ngoc Cong Duong ◽  
Tien Nguyen Le Bao ◽  
Phuong Thi Lan Nguyen ◽  
Thanh Vo Van ◽  
Toi Phung Lam ◽  
...  
Keyword(s):  
Psychological Distress ◽  
General Population ◽  
Psychological Health ◽  
Online Survey ◽  
Survey Study ◽  
Limited Information ◽  
Cross Sectional Survey ◽  
Cross Sectional ◽  
Event Scale ◽  
The Impact

BACKGROUND The first nationwide lockdown due to the COVID-19 pandemic was implemented in Vietnam from April 1 to 15, 2020. Nevertheless, there has been limited information on the impact of COVID-19 on the psychological health of the public. OBJECTIVE This study aimed to estimate the prevalence of psychological issues and identify the factors associated with the psychological impact of COVID-19 during the first nationwide lockdown among the general population in Vietnam. METHODS We employed a cross-sectional study design with convenience sampling. A self-administered, online survey was used to collect data and assess psychological distress, depression, anxiety, and stress of participants from April 10 to 15, 2020. The Impact of Event Scale-Revised (IES-R) and the Depression, Anxiety, and Stress Scale-21 (DASS-21) were utilized to assess psychological distress, depression, anxiety, and stress of participants during social distancing due to COVID-19. Associations across factors were explored using regression analysis. RESULTS A total of 1385 respondents completed the survey. Of this, 35.9% (n=497) experienced psychological distress, as well as depression (n=325, 23.5%), anxiety (n=195, 14.1%), and stress (n=309, 22.3%). Respondents who evaluated their physical health as average had a higher IES-R score (beta coefficient [B]=9.16, 95% CI 6.43 to 11.89), as well as higher depression (B=5.85, 95% CI 4.49 to 7.21), anxiety (B=3.64, 95% CI 2.64 to 4.63), and stress (B=5.19, 95% CI 3.83 to 6.56) scores for DASS-21 than those who rated their health as good or very good. Those who self-reported their health as bad or very bad experienced more severe depression (B=9.57, 95% CI 4.54 to 14.59), anxiety (B=7.24, 95% CI 3.55 to 10.9), and stress (B=10.60, 95% CI 5.56 to 15.65). Unemployment was more likely to be associated with depression (B=3.34, 95% CI 1.68 to 5.01) and stress (B=2.34, 95% CI 0.84 to 3.85). Regarding worries about COVID-19, more than half (n=755, 54.5%) expressed concern for their children aged <18 years, which increased their IES-R score (B=7.81, 95% CI 4.98 to 10.64) and DASS-21 stress score (B=1.75, 95% CI 0.27 to 3.24). The majority of respondents (n=1335, 96.4%) were confident about their doctor’s expertise in terms of COVID-19 diagnosis and treatment, which was positively associated with less distress caused by the outbreak (B=–7.84, 95% CI –14.58 to –1.11). CONCLUSIONS The findings highlight the effect of COVID-19 on mental health during the nationwide lockdown among the general population in Vietnam. The study provides useful evidence for policy decision makers to develop and implement interventions to mitigate these impacts. CLINICALTRIAL

scholarly journals A Call for a Reform of the Influenza Immunization Program in Mexico: Epidemiologic and Economic Evidence for Decision Making

Vaccines ◽  
2021 ◽  
Vol 9 (3) ◽  
pp. 286
Author(s):  
Roberto Tapia-Conyer ◽  
Miguel Betancourt-Cravioto ◽  
Alejandra Montoya ◽  
Jorge Abelardo Falcón-Lezama ◽  
Myrna María Alfaro-Cortes ◽  
...  
Keyword(s):  
Cost Effectiveness ◽  
Influenza Vaccination ◽  
Vaccination Coverage ◽  
Decision Analytic Model ◽  
Limited Information ◽  
Vaccination Policy ◽  
Cross Sectional ◽  
Influenza Case ◽  
One Year ◽  
Effectiveness Studies

Limited information is available to determine the effectiveness of Mexico’s national influenza vaccination guidelines and inform policy updates. We aim to propose reforms to current influenza vaccination policies based on our analysis of cost-effectiveness studies. This cross-sectional epidemiological study used influenza case, death, discharge and hospitalization data from several influenza seasons and applied a one-year decision-analytic model to assess cost-effectiveness. The primary health outcome was influenza cases avoided; secondary health outcomes were influenza-related events associated with case reduction. By increasing vaccination coverage to 75% in the population aged 12–49 years with risk factors (diabetes, high blood pressure, morbid obesity, chronic renal failure, asthma, pregnancy), and expanding universal vaccination coverage to school-aged children (5–11 years) and adults aged 50–59 years, 7142–671,461 influenza cases; 1–15 deaths; 7615–262,812 healthcare visits; 2886–154,143 emergency room admissions and 2891–97,637 hospitalizations could be prevented (ranges correspond to separate age and risk factor groups), with a net annual savings of 3.90 to 111.99 million USD. Such changes to the current vaccination policy could potentially result in significant economic and health benefits. These data could be used to inform the revision of a vaccination policy in Mexico with substantial social value.

scholarly journals A cross-sectional study reporting concussion exposure, assessment and management in Western Australian general practice

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Elizabeth Thomas ◽  
HuiJun Chih ◽  
Belinda Gabbe ◽  
Melinda Fitzgerald ◽  
Gill Cowen
Keyword(s):  
General Practice ◽  
Cross Sectional Study ◽  
Limited Information ◽  
Sectional Study ◽  
Cross Sectional ◽  
Usual Practice ◽  
Exact Test ◽  
Concussion Management ◽  
Practice Knowledge ◽  
Western Australian

Abstract Background General Practitioners (GPs) may be called upon to assess patients who have sustained a concussion despite limited information being available at this assessment. Information relating to how concussion is actually being assessed and managed in General Practice is scarce. This study aimed to identify characteristics of current Western Australian (WA) GP exposure to patients with concussion, factors associated with GPs’ knowledge of concussion, confidence of GPs in diagnosing and managing patients with concussion, typical referral practices and familiarity of GPs with guidelines. Methods In this cross-sectional study, GPs in WA were recruited via the RACGP WA newsletter and shareGP and the consented GPs completed an electronic survey. Associations were performed using Chi-squared tests or Fisher’s Exact test. Results Sixty-six GPs in WA responded to the survey (response rate = 1.7%). Demographics, usual practice, knowledge, confidence, identification of prolonged recovery as well as guideline and resource awareness of GPs who practised in regional and metropolitan areas were comparable (p > 0.05). Characteristics of GPs were similar between those who identified all symptoms of concussion and distractors correctly and those who did not (p > 0.05). However, 84% of the respondents who had never heard of concussion guidelines were less likely to answer all symptoms and distractors correctly (p = 0.039). Whilst 78% of the GPs who were confident in their diagnoses had heard of guidelines (p = 0.029), confidence in managing concussion was not significantly associated with GPs exposure to guidelines. It should be noted that none of the respondents correctly identified signs of concussion and excluded the distractors. Conclusions Knowledge surrounding concussion guidelines, diagnosis and management varied across GPs in WA. Promotion of available concussion guidelines may assist GPs who lack confidence in making a diagnosis. The lack of association between GPs exposure to guidelines and confidence managing concussion highlights that concussion management may be an area where GPs could benefit from additional education and support.

Sign in / Sign up

Export Citation Format

Share Document