Hands-On Machine Learning for Algorithmic Trading

Receiver operating characteristics and the area under the curve

The receiver operating characteristics (ROC) curve allows us to visualize, organize, and select classifiers based on their performance. It computes all the combinations of true positive rates (TPR) and false positive rates (FPR) that result from producing predictions using any of the predicted scores as a threshold. It then plots these pairs on a square, the side of which has a measurement of one in length.

A classifier that makes random predictions (taking into account class imbalance) will on average yield TPR and FPR that are equal so that the combinations will lie on the diagonal, which becomes the benchmark case. Since an underperforming classifier would benefit from relabeling ...