Pitney Bowes MapInfo Vertical Mapper - Manual (Page 182)

Point Aggregation with Statistics
180
Vertical Mapper 3.7
• Number of Points Aggregated: This value is the total number of points selected for
aggregation. It does not include the points processed in the Coincident Point Handling step.
• Standard Deviation: This value is the degree of dispersion between the mean of the values for
those points selected for aggregation.
• % Normalized Coeff. of Variation: The coefficient of variation is the standard deviation divided
by the average expressed as a percentage (multiplied by 100).
The value is dimensionless and indicates data dispersion, especially when the average is not
close to zero. A “normalized” coefficient of variation is calculated using the standard deviation
divided by the difference between the average and the minimum value of the complete data set,
not just the data in the local aggregated region. The coefficient can still be large when the
average value is close to the minimum value, but the normalized coefficient is representative of
the original range of values and not their absolute value.
The Aggregation Distance box—enables you to specify the distance used to group points for
aggregation.
This distance is defined differently for each of the three aggregation techniques. For the Cluster
Density and Forward Stepping techniques, aggregation distance is defined by the radius of a user-
specified circular search area centred on each aggregation cell. For the Square Cell technique,
Aggregation Distance is defined by the width of the square aggregation cell.
The Create Regions Table check box—enables you to build a table of the regions used to group the
point data selected for aggregation.
This option enables you to visually inspect the results of the aggregation process. The order in which
these regions are created is the same order in which the point file was aggregated. By opening a
Browser window of the region file and choosing each record in the list one at a time, you can see
how the regions are processed in the Map window.
Note You can use the created regions to produce a MapInfo Professional coloured thematic map
where each region is thematically shaded according to one of the computed statistical
values.
The Create STD Ellipse Table check box—enables you to build a table of standard deviation
ellipses. For more information, see
Building a Table of Standard Deviation Ellipses on page 183 .
Forward Stepping Aggregation
Due to its speed and effectiveness, the Forward Stepping Aggregation technique is an appropriate
method for any general aggregation application. This technique is useful when your data has a truly
random distribution or when the other two techniques are not appropriate.
This aggregation process aggregates points by moving through the data set from left to right and
then top to bottom. The process begins by sorting the data points into rows that are generally three
times the aggregation distance. This sorting is performed to determine where the aggregation will
begin each left-to-right swath. There are no settings you can use to alter this process. Beginning
with the data points in the upper left (northwest) corner of the data set, a circular search radius is
created as specified by the Aggregation Distance setting. All data points that fall inside this search
radius are selected and flagged. This prevents these data points from being aggregated to another
location. The geocenter of the selected points is then determined. This becomes the location of the