Similar to other predict methods, this functions predicts fitted values, logits,
coefficients and more from a fitted "glmnet" object.
# S3 method for glmnet
predict(object, newx, s = NULL,
type=c("link","response","coefficients","nonzero","class"), exact = FALSE, offset, ...)
# S3 method for glmnet
coef(object,s=NULL, exact=FALSE, ...)Fitted "glmnet" model object.
Matrix of new values for x at which predictions are
to be made. Must be a matrix; can be sparse as in Matrix
package. This argument is not used for type=c("coefficients","nonzero")
Value(s) of the penalty parameter lambda at which
predictions are required. Default is the entire sequence used to
create the model.
Type of prediction required. Type "link" gives the
linear predictors for "binomial", "multinomial",
"poisson" or "cox" models; for "gaussian"
models it gives the fitted
values. Type "response" gives the fitted probabilities for
"binomial" or "multinomial", fitted mean for
"poisson" and the fitted relative-risk for "cox"; for "gaussian"
type "response" is equivalent to type "link". Type
"coefficients" computes the coefficients at the requested
values for s. Note that for
"binomial" models, results are returned only for the class
corresponding to the second level of the factor response.
Type "class" applies only to
"binomial" or "multinomial" models, and produces the
class label corresponding to the maximum probability. Type
"nonzero" returns a list of the indices of the nonzero
coefficients for each value of s.
This argument is relevant only when predictions are made at
values of s (lambda) different from those used in the fitting
of the original model. If exact=FALSE (default), then the predict function uses linear interpolation
to make predictions for values of s (lambda) that do not coincide with
those used in the fitting algorithm. While this is often a good
approximation, it can sometimes be a bit coarse.
With exact=TRUE, these different values of s are merged
(and sorted) with
object$lambda, and the model is refit before predictions are
made. In this case, it is strongly advisable to supply the original
data x= and y= as additional
named arguments to predict() or coef().
The workhorse predict.glmnet() needs to update the model,
and expects the data used to create it to be around.
Although it will often work fine without supplying these additional arguments, it will likely break when
used inside a nested sequence of function calls.
If additional arguments such as weights and
offset were used, these should be included by name as well.
If an offset is used in the fit, then one must be
supplied for making predictions (except for
type="coefficients" or type="nonzero")
This is the mechanism for passing arguments like
x= when exact=TRUE; seeexact argument.
The object returned depends on type.
The shape of the objects returned are different for
"multinomial" objects. This function actually calls
NextMethod(),
and the appropriate predict method is invoked for each of the three
model types. coef(...) is equivalent to predict(type="coefficients",...)
Friedman, J., Hastie, T. and Tibshirani, R. (2008) Regularization Paths for Generalized Linear Models via Coordinate Descent, http://www.stanford.edu/~hastie/Papers/glmnet.pdf Journal of Statistical Software, Vol. 33(1), 1-22 Feb 2010 http://www.jstatsoft.org/v33/i01/ Simon, N., Friedman, J., Hastie, T., Tibshirani, R. (2011) Regularization Paths for Cox's Proportional Hazards Model via Coordinate Descent, Journal of Statistical Software, Vol. 39(5) 1-13 http://www.jstatsoft.org/v39/i05/
glmnet, and print, and coef methods, and cv.glmnet.
# NOT RUN {
x=matrix(rnorm(100*20),100,20)
y=rnorm(100)
g2=sample(1:2,100,replace=TRUE)
g4=sample(1:4,100,replace=TRUE)
fit1=glmnet(x,y)
predict(fit1,newx=x[1:5,],s=c(0.01,0.005))
predict(fit1,type="coef")
fit2=glmnet(x,g2,family="binomial")
predict(fit2,type="response",newx=x[2:5,])
predict(fit2,type="nonzero")
fit3=glmnet(x,g4,family="multinomial")
predict(fit3,newx=x[1:3,],type="response",s=0.01)
# }
Run the code above in your browser using DataLab