The feature view does not contain any statistics, as it is simply an interface consisting of a number of features and any transformation functions applied to those features.

However, training data can have descriptive statistics over it computed by HSFS. Descriptive statistics for training data is important for model monitoring, as it can enable model monitoring. If you compute the same descriptive statistics over windows of input features to models, you can help determine when there is a significant change in the distribution of an input feature, so-called feature shift.