sso was applied to the olive oil content data. The result was

t with the RLR model for this data set.

Fig. 4.21. The Lasso model constructed for the olive oil content data.

e elastic net linear regression algorithm

ic net linear regression algorithm (ELR) introduces two Lagrange

ou and Hastie, 2005] as shown below, where ߣଵ and ߣ૛ stand for

ive constants. It can be seen that ELR is a shrinkage mixture, i.e.,

e of RLR represented by the ߣଵ term and Lasso regression

ed by the ߣ૛ term.

ሺܡെ܆ܟሻ^௧ሺܡെ܆ܟሻ൅ߣଵሺܟ^௧ܟെܿሻ൅ߣ૛෍|ݓ௜|

ௗ

௜ୀ଴

(4.46)

lmnet package can also be used for ELR, in which the alpha

r is set to a value between zero and one, for instance, 0.5,

glmnet(x,y,alpha=0.5,lambda)

e 4.22 shows how the regression coefficients evolve in a learning

of a ELR model constructed for the olive oil content data. The

very similar to the RLR model and the Lasso model which were

ated above.