Pattern and Process: Application of local and space-time spatial statistics to epidemiological data.
Summary
Lab
- Snow Cholera data
- Distribution of cases
- Distribution of pumps
- Thiessen polygons around pumps
- Gi(d) local spatial statistic
- Trinidad malaria cases
- Spatial distribution of cases
- Temporal distribution of cases
- Cases by source
- Cases by parasite species
- k-NN space-time statistic
- Encephalitis in Illinois and the Tiger Mosquito
- Ticks on deer around a known Lyme focus
- Distribution of infested deer
- Distribution of uninfested deer
- changes of distribution over time
- Second order statistic
There will be a Class exercise associated with the lecture, using Trinidad
malaria data.
- Mapping:
- Changing distribution of cases by country of origin over time (e.g.,
Nigeria, India or Guyana).
- Grouping of all cases within a given area (e.g., 25 km2), and then
mapping clusters of cases (e.g. as circles with size related to number of
cases). Maps can then be made for a variety of categories (temporal, origin,
species, etc.).
- Spatial statistics:
- Moran's I for the distribution of cases for the three parasite species.
- k-NN for cases from selected continents or countries.
Readings
Page by Andy Long.
Comments appreciated.
aelon@sph.umich.edu