Effective Geographic Sample Size in the Presence of Spatial Autocorrelation

Griffith, Daniel A.

doi:10.1111/j.1467-8306.2005.00484.x

Cited by 163 publications

(143 citation statements)

References 36 publications

Supporting

Mentioning

114

Contrasting

Unclassified

Order By: Relevance

“…The distance separating the two points, noted ij d , can be calculated using the Pythagorean theorem (Equation (1)) 3 .…”

Section: Building An Appropriate Weights Matrixmentioning

confidence: 99%

See 1 more Smart Citation

Modeling Spatial Data Pooled over Time: Schematic Representation and Monte Carlo Evidences

Dubé¹,

Legros

2015

TEL

View full text Add to dashboard Cite

The spatial autocorrelation issue is now well established, and it is almost impossible to deal with spatial data without considering this reality. In addition, recent developments have been devoted to developing methods that deal with spatial autocorrelation in panel data. However, little effort has been devoted to dealing with spatial data (cross-section) pooled over time. This paper endeavours to bridge the gap between the theoretical modeling development and the application based on spatial data pooled over time. The paper presents a schematic representation of how spatial links can be expressed, depending on the nature of the variable, when combining the spatial multidirectional relations and temporal unidirectional relations. After that, a Monte Carlo experiment is conducted to establish the impact of applying a usual spatial econometric model to spatial data pooled over time. The results suggest that neglecting the temporal dimension of the data generating process can introduce important biases on autoregressive parameters and thus result in the inaccurate measurement of the indirect and total spatial effect related to the spatial spillover effect.

show abstract

“…The distance separating the two points, noted ij d , can be calculated using the Pythagorean theorem (Equation (1)) 3 .…”

Section: Building An Appropriate Weights Matrixmentioning

confidence: 99%

“…Its complexity explains why spatial autocorrelation has received such attention since spatial data are now widely available and used. Spatial autocorrelation among residuals of a statistical model can have various consequences on estimated coefficients and variances, depending on sample size [3]- [5].…”

Section: Introductionmentioning

confidence: 99%

Modeling Spatial Data Pooled over Time: Schematic Representation and Monte Carlo Evidences

Dubé¹,

Legros

2015

TEL

View full text Add to dashboard Cite

show abstract

“…With the recent advances and improvements in GIS applications and the ensuing infusion of other associated statistical packages into geographical software, models have been developed to account for spatial autocorrelation (Cliff and Ord, 1973;Getis, 1992, 1998;Bailey and Gatrell, 1995;Griffith, 2006;Yu and Wei, 2008). The application of these techniques specifically for studies of mosquito-borne diseases has attracted increasing attention (Li et al, 2008;Feng et al, 2011;Impoinvil et al, 2011).…”

Section: Introductionmentioning

confidence: 99%

“…Map of Australia with the study area highlighted using dark and light shades. might lead to unreliable estimates and potentially misleading inferences (Cliff and Ord, 1973;Anselin and Getis, 1992;Bailey and Gatrell, 1995;Anselin and Bera, 1998;Griffith, 2006;Yu and Wei, 2008).…”

Section: Introductionmentioning

confidence: 99%

Spatial analysis of risk factors for transmission of the Barmah Forest virus in Queensland, Australia

2013

View full text Add to dashboard Cite

Abstract. Barmah Forest virus (BFV) disease is the second most common mosquito-borne disease in Australia but few data are available on the risk factors. We assessed the impact of spatial climatic, socioeconomic and ecological factors on the transmission of BFV disease in Queensland, Australia, using spatial regression. All our analyses indicate that spatial lag models provide a superior fit to the data compared to spatial error and ordinary least square models. The residuals of the spatial lag models were found to be uncorrelated, indicating that the models adequately account for spatial and temporal autocorrelation. Our results revealed that minimum temperature, distance from coast and low tide were negatively and rainfall was positively associated with BFV disease in coastal areas, whereas minimum temperature and high tide were negatively and rainfall was positively associated with BFV disease (all P-value <0.05). The study demonstrates that BFV disease is more densely distributed in coastal areas and is influenced by climatic and ecological factors. The spatial analytical approach used in this study may have ramifications in the planning and implementation of BFV disease prevention and control programmes.

show abstract

“…We instead estimate n * using an approach based on Griffith (2005), an approach that does not require computing V −1 directly:…”

mentioning

confidence: 99%

Characterizing biospheric carbon balance using CO<sub>2</sub> observations from the OCO-2 satellite

Miller¹,

Michalak²,

Yadav³

et al. 2017

Preprint

View full text Add to dashboard Cite

Abstract.NASA's Orbiting Carbon Observatory-2 (OCO-2) satellite launched in summer of 2014. Its observations could allow scientists to constrain CO 2 fluxes across regions or continents that were previously difficult to monitor. This study explores an initial step toward that goal; we evaluate the extent to which current OCO-2 observations can detect patterns in biospheric CO 2 fluxes and constrain monthly CO 2 budgets. Our goal is to guide top-down, inverse modeling studies and identify areas 5 for future improvement. We find that uncertainties and biases in the individual OCO-2 observations are comparable to the atmospheric signal from biospheric fluxes, particularly during northern hemisphere winter when biospheric fluxes are small. A series of top-down experiments indicate how these errors affect our ability to constrain monthly biospheric CO 2 budgets. We are able to constrain budgets for between two and four global regions using OCO-2 observations, depending on the month, and we can constrain CO 2 budgets at the regional level (i.e., smaller than seven global biomes) in only a handful of cases (16% of 10 all regions and months). The potential of the OCO-2 observations, however, is greater than these results might imply. A set of synthetic data experiments suggests that observation or retrieval errors have a salient effect. Advances in retrieval algorithms and to a lesser extent atmospheric transport modeling will improve the results. In the interim, top-down studies that use current satellite observations are best-equipped to constrain the biospheric carbon balance across only continental or hemispheric regions.

show abstract

Effective Geographic Sample Size in the Presence of Spatial Autocorrelation

Cited by 163 publications

References 36 publications

Modeling Spatial Data Pooled over Time: Schematic Representation and Monte Carlo Evidences

Modeling Spatial Data Pooled over Time: Schematic Representation and Monte Carlo Evidences

Spatial analysis of risk factors for transmission of the Barmah Forest virus in Queensland, Australia

Characterizing biospheric carbon balance using CO<sub>2</sub> observations from the OCO-2 satellite

Contact Info

Product

Resources

About

Effective Geographic Sample Size in the Presence of Spatial Autocorrelation

Cited by 163 publications

References 36 publications

Modeling Spatial Data Pooled over Time: Schematic Representation and Monte Carlo Evidences

Modeling Spatial Data Pooled over Time: Schematic Representation and Monte Carlo Evidences

Spatial analysis of risk factors for transmission of the Barmah Forest virus in Queensland, Australia

Characterizing biospheric carbon balance using CO&lt;sub&gt;2&lt;/sub&gt; observations from the OCO-2 satellite

Contact Info

Product

Resources

About

Characterizing biospheric carbon balance using CO<sub>2</sub> observations from the OCO-2 satellite