Data Integration Model for Air Quality: A Hierarchical Approach to the Global Estimation of Exposures to Ambient Air Pollution

Shaddick, Gavin; Thomas, Matthew L.; Jobling, Amelia; Bräuer, Michael; Donkelaar, Aaron van; Burnett, Rick; Chang, Howard Y.; Cohen, Aarón; Dingenen, Rita Van; Dora, Carlos; Gumy, Sophie; Liu, Yang; Martin, Randall V.; Waller, Lance A.; West, J. Jason; Zidek, James V.; Prüss‐Üstün, Annette

doi:10.48550/arxiv.1609.00141

Cited by 1 publication

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These three models will be sufficient for demonstrating our proposed workflow, but this is a smaller network of models than we would use for a comprehensive analysis of the PM 2.5 data. Shaddick et al (2017), for example, also consider smaller regions, countrylevel variation, and a spatial model for the varying coefficients. Further calibration covariates can also be included.…”

Section: Exploratory Data Analysis Goes Beyond Just Plotting the Datamentioning

confidence: 99%

“…exhibited by Models 2 and 3 indicate that the univariate predictive distributions are too broad compared to the data, which suggests that further modeling will be necessary to accurately reflect the uncertainty. One possibility would be to further sub-divide the super-regions to better capture within-region variability (Shaddick et al, 2017).…”

Section: How Did We Do? Posterior Predictive Checks Are Vital For Mod...mentioning

confidence: 99%

“…In order to better discuss the ways visualization can aid a statistical workflow we consider a particular problem, the estimation of human exposure to air pollution from particulate matter measuring less than 2.5 microns in diameter (PM 2.5 ). Exposure to PM 2.5 is linked to a number of poor health outcomes, and a recent report estimated that PM 2.5 is responsible for three million deaths worldwide each year (Shaddick et al, 2017).…”

Section: Introduction and Running Examplementioning

confidence: 99%

“…For our running example, we use the data from Shaddick et al (2017), aggregated to the city level, to estimate ambient PM 2.5 concentration across the world. The statistical problem is that we only have direct measurements of PM 2.5 from a sparse network of 2980 ground monitors with heterogeneous spatial coverage (Figure 1a).…”

Section: Introduction and Running Examplementioning

confidence: 99%

“…The aim of this analysis is to build a predictive model of PM 2.5 with appropriately calibrated prediction intervals. We will not attempt a full analysis of this data, which was undertaken by Shaddick et al (2017). Instead, we will focus on three simple, but plausible, models for the data in order to show how visualization can be used to help construct, sense-check, compute, and evaluate these models.…”

Section: Introduction and Running Examplementioning

confidence: 99%

See 4 more Smart Citations

Visualization in Bayesian Workflow

Gabry

Simpson

Vehtari

et al. 2019

Journal of the Royal Statistical Society Series A: Statistics in Society

742

686

View full text Add to dashboard Cite

Bayesian data analysis is about more than just computing a posterior distribution, and Bayesian visualization is about more than trace plots of Markov chains. Practical Bayesian data analysis, like all data analysis, is an iterative process of model building, inference, model checking and evaluation, and model expansion. Visualization is helpful in each of these stages of the Bayesian workflow and it is indispensable when drawing inferences from the types of modern, high dimensional models that are used by applied researchers.

show abstract

Section: Exploratory Data Analysis Goes Beyond Just Plotting the Datamentioning

confidence: 99%

Section: How Did We Do? Posterior Predictive Checks Are Vital For Mod...mentioning

confidence: 99%

Section: Introduction and Running Examplementioning

confidence: 99%

Section: Introduction and Running Examplementioning

confidence: 99%

Section: Introduction and Running Examplementioning

confidence: 99%

See 3 more Smart Citations