Pitney Bowes MapInfo Vertical Mapper - Manual (Page 185)

Chapter 10: Aggregating Data
User Guide
183
There are a total of 330 points in the sample data set shown in the previous figure. After using the
cluster density aggregator, you can see that the data is clustered in about ten separate areas. The
more randomly distributed points lying outside these clustered zones (circles) are also aggregated,
but with significantly fewer points included. Although many circles overlap, the degree of overlap is
significantly less than when the forward stepping method is used.
Square Cell Aggregation
The Square Cell Aggregation technique is typically used when values are required that represent
specific areas, for example, a density map of the number of new housing units per square kilometre,
or when you want to avoid overlapping aggregation regions.
This method divides the area covered by the point file into adjacent squares determined by the
aggregation distance. The points that fall inside any of these squares are aggregated to a new point
created at the geocentre of the aggregated points (not at the centre of the square). As with the
previous two techniques, the specified statistical information is then attached as attributes to the new
aggregated point.
An example of data points aggregated using the Square Cell Aggregation technique.
Although there is no overlap of the aggregation regions in the figure above, points have been
aggregated inappropriately in several areas. Therefore, the best results require a certain degree of
overlap.
Building a Table of Standard Deviation Ellipses
Standard deviation is a measure of dispersion in point patterns. Typically, it measures dispersal in
terms of a circle around the mean centre. The circular model, however, takes no account of the fact
that spread may be different in different directions.
The standard deviation ellipse summarizes dispersion in a point pattern in terms of an ellipse rather
than a circle. The ellipse is centred on the mean centre, with its long axis in the direction of the
maximum dispersion and its short axis in the direction of the minimum dispersion. The axis of