Validating National Kriging Exposure Estimation
Author Information
Author(s): Adam A. Szpiro, Lianne Sheppard, Paul D. Sampson, Sun-Young Kim
Primary Institution: University of Washington
Hypothesis
Is national scale daily kriging a feasible method for estimating ambient air pollution exposure?
Conclusion
The proposed methodology for national kriging is appealing but lacks convincing evidence of reliability.
Supporting Evidence
- The authors argue that national kriging is preferable to regional kriging.
- Automated variogram estimation is claimed to be better than manual estimation.
- Cross-validation statistics such as prediction error and RMSS were discussed.
Takeaway
The study looks at a method to estimate air pollution levels across the country, but the results aren't strong enough to trust completely.
Methodology
The study discusses the use of national scale daily kriging to estimate PM10 exposure based on EPA measurements and evaluates its performance through cross-validation statistics.
Potential Biases
The study may have bias risks due to reliance on limited cross-validation sites.
Limitations
The evidence supporting the methodology is not convincing, and some necessary statistics were not reported.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website