Geoscience Reference
In-Depth Information
Figure 3.11.
Spatio-temporal analysis of cities' economic
specialization in US [PAU 10]
This type of analysis is extremely rich in revealing the evolution structures from
a multivariate description of the entities. It exploits the fact that the temporal
dimension is explicit (time “unfolded”). The evolution map can then be represented
(“support” space). They can be conducted on the contingency tables or
heterogeneous tables referring to a diversity of thematics, such as, for example, a set
of basins described by flow measurements over time.
3.3.2.
Identifying a priori defined types of change to classify evolutions
The previous example was based upon easily formalized evolutions in a
statistical table: the objects are well identified and measurements easily repeatable in
time. In the case that follows the situation is a little different. It concerns assessing
the change of a set of composite objects, the construction of which is the result of a
multi-criteria analysis with several steps. This is the case of the employment centers
of the urban area of Paris. They have been defined on the basis of statistical data at
the level of the municipality from the year 2000, with a method combining a multi-
criteria approach to identify the elementary entities (municipalities) that can be
considered as the core of the centers and a spatial analysis to then aggregate them
into employment centers [BER 11a]. The multi-criteria approach has called up
indicators describing, on the one hand, the concentration of jobs and, on the other
hand, the attraction exerted by the municipalities on the rest of the territory. The
spatial analysis method has called up indicators to identify the shape of interactions
