Which is a characteristic feature of lasso regression?

Lasso regression is characterized by its ability to perform both variable selection and regularization, which helps to enhance the model's interpretability while preventing overfitting. One of the key characteristics of lasso regression is that it can effectively reduce the number of predictors in a model. This is achieved through the use of L1 regularization, which adds a penalty equivalent to the absolute value of the magnitude of coefficients. This penalty encourages some coefficients to be exactly zero, effectively removing those predictors from the model.

This feature is particularly valuable in situations where there are many predictors, as it simplifies the model, helps to identify the most important variables, and thus can lead to a more robust model that generalizes better on new data. The ability to perform feature selection makes lasso regression a popular choice in scenarios where interpretability and the elimination of irrelevant features are important considerations.