You can perform out-of-sample prediction by estimating the sample regression function for
A)all of the data,using the results to predict outcomes for all of the data,and comparing the predicted outcomes to the actual outcomes.
B)a subset of the data,using the results to predict outcomes for all of the data,and comparing the predicted outcomes to the actual outcomes.
C)all of the data,using the results to predict outcomes for a subset of the data,and comparing the predicted outcomes to the actual outcomes.
D)a subset of the data,using the results to predict outcomes for the subset not included in the regression,and comparing the predicted outcomes to the actual outcomes.

Question

Quizplus · Accepted Answer

Out-of-sample prediction involves using the results of a regression model to predict outcomes for data not included in the original regression. The only choice that involves predicting outcomes for a subset not included in the regression is D. Choices A, B, and C all involve using the regression results to predict outcomes for all or some of the data that were included in the original regression, which is not out-of-sample prediction. Therefore, D is the best choice.

You Can Perform Out-Of-Sample Prediction by Estimating the Sample Regression