Understanding the Importance of Resampling in Machine Learning

To underestimate the test error rate

To fit a model of interest to samples formed from the training subset (correct)

To directly obtain a large designated test set

To separate a dataset into training and testing subsets

To divide the available set of samples into two parts

To estimate test error from the resulting validation set error (correct)

To quantify the uncertainty associated with a given estimator

To identify the method that results in the highest test error

Data sets obtained by fitting a model of interest to samples formed from the training subset

Data sets used to directly estimate the prediction error

Data sets created by sampling with replacement from the original data set (correct)

Data sets created by dividing the available samples into training and validation sets

<p>To estimate standard deviation and bias of parameter estimates</p> Signup and view all the answers

<p>To provide an estimate of the standard error of a coefficient</p> Signup and view all the answers

<p>By dividing data randomly into training and validation subsets</p> Signup and view all the answers