Aggregation

When the Aggregation option is selected, the statistics are computed in terms of predicted frequencies. In models with categorical response variables, statistics can be computed in terms of the raw data or for aggregated frequency counts. In the binomial case, the response variable has two possible values: 0 or 1. Accordingly, predicted values should be computed that fall in the range from 0 to 1. If the Aggregation checkbox is selected, Statistica® considers the aggregated (tabulated) data set. In that case, the response variable is a frequency count, reflecting the number of observations that fall into the respective categories. As an example consider the following data:

x1	x2	Y
1.5	2	1
2.5	1	0
2.5	1	1
1.5	2	0
2.5	1	1
1.5	2	1

After aggregation, the data is represented as follows:

x1	x2	y1	y0	total
1.5	2	2	1	3
2.5	1	2	1	3

Selecting the Aggregation checkbox also affects the computation (and display) of predicted and residual values.