Abstract
The e xercise of ident if ying, quan tifying and int erpreting spatio-temporal pa tt e rns d epends h ighly on the method used to divide a given lands cape into d iscrete territ ori a l units (reg ions). Given that a nisotropy is a common chara cte ristic observed a long s patial su rfaces, it should be cl ea rly as sumed t ha t t he dist r ibut ion of agent s ( events , obje ct s, indi vidual s, et c .) i s heterog eneous. Th is non-stationarity in the d ata c auses a ggregation proble ms rel ated to what is common ly known a s ecological fallacy. In temporal dyn amic analysis o f s patial data, when inf ormation is colle cte d and a ggregated for dist inct s ets of spatial units - distinct g eometrical s chemes - then the Modif iable Ar eal Uni t Problem (MA UP) occurs. Thi s article demonst rates how the se issue s may be tackl ed thr ough the us e of an cillary geog raphical info rmation (control zones). The motivation of the present research was the need to crea te, for s mall area s, cohere nt databas es from census data c ollec ted for d istinct ti me -stamps, whi ch will, for the first ti me, allow the dyn amic study o f demographic p atterns for a s peci fic study-a rea - the l ocal coun cil o f Tom ar. Information is agg regated for existing census trac ts ("sub-secçõe s") for 2001 and 20 11. The ha rm onisat ion of the two d atasets is not pos sible s ince the re is no c ommon cod ing syste m which would allow aggr egation fo r co mmon z ones, bec ause of g eometric i nc onsis tenci es. Us ing ve ctor a nalysis and trans format ion, da ta from both c ensus ex ercises was re-allocated for common regi ons, based on a weighting scheme ; this in turn was produced using control z ones. It is a rgued that the methodology us ed minimises the e rror a ssociate d with asymmetr ic aggrega tion of spa ti a l da ta. Finally, the a rticle con cludes with an example o f the potential us e of the proposed methodology.
| Original language | Unknown |
|---|---|
| Title of host publication | XIX Jornadas de Classificação e Análise de Dados |
| Editors | L. Grilo |
| Place of Publication | Tomar |
| Publisher | Instituto Politécnico de Tomar |
| Pages | 159-165 |
| ISBN (Print) | 978-972-9473-60-9 |
| Publication status | Published - 1 Jan 2012 |