Statistical Analysis (3)
A more compact expression is:
At this point we can estimate expected value and variance of the weights:
Where ?2 is the variance of y; correlation analysis is now easy.
Notes:
The analysis gives as a result the correlation between the real data and the values obtained by the metrics.
If the metric under validation is a complex metric, not only the correlation with the total metric is given, but also the correlation with all the single terms of the metric.