Bayesian areal wombling on simulated disease maps whose risk changes only by degrees
A research poster from the School of Underlying Conditions fits a Bayesian areal wombling rule to 3,000 simulated disease maps of one synthetic region, whose relative risk changes either as a step or as a linear ramp up to 16 areal units wide. The method recovers a true step in 96% of maps, but still declares a boundary on most ramps up to four units wide, placing it anywhere across the ramp, so that on graded ground the mapped line records where the rule fell rather than an edge in the surface.
| Authors | Brynjar Obiora , Vigdis Emenike , Bram Ntuli |
|---|---|
| School | School of Underlying Conditions |
| Output type | Research poster |
| Published | |
| DOI | 10.5555/slop.r3zao7 |
| Pages | 1 |
| Version | 1.0 |
| Licence | CC BY 4.0 |
Cites
Cite as
@misc{slop_r3zao7,
author = {Brynjar Obiora and Vigdis Emenike and Bram Ntuli},
title = {A Line Somewhere on the Ramp: Bayesian areal wombling on simulated disease maps whose risk changes only by degrees},
year = {2026},
publisher = {Slop University},
doi = {10.5555/slop.r3zao7},
url = {https://slop.university/outputs/slop-poster-how-often-bayesian-r3zao7/},
version = {1.0},
note = {Research poster},
}