Tobit regression with elastic net in python

Tobit regression is a form of censored regression that can handle a mix of left- and right-censored (and, of course, uncensored) observations of the target variable.

The R package censReg guided much of the implementation.

• p_censor_left The censor point on the left side.

• p_censor_right The censor point on the right side.

• C The regularization parameter, a larger C represents a heavier degree of penalty.

• alpha The alpha governs the balance between the L1 and L2 regularization terms.