Review Feature Engineering

Let's briefly review what we learned in the feature engineering lesson. Which of the following statements are true?

When using a machine learning model for alpha combination, the input columns must be alpha factors.

A sector feature is helpful because it allows the model to learn whether different factors are relevant for different sectors/industries.

When choosing the targets/labels of a machine learning model for alpha combination, raw stock return values are recommended.

Features can help a machine learning model identify conditions when other features or factors have are more or less predictive of stock returns.

Market dispersion is calculated as the standard deviation of normalized values of several common market indices.

SOLUTION:

A sector feature is helpful because it allows the model to learn whether different factors are relevant for different sectors/industries.
Features can help a machine learning model identify conditions when other features or factors have are more or less predictive of stock returns.