Cross-validation (CV) is a popular strategy for model selection. The main idea behind CV is to split the data one or several times so that each split is used once as a validation set and the remainder as a training set: part of the data (the training sample) is used to train the algorithm, and the remaining part (the validation sample) is used for estimating the risk of the algorithm. Then, CV selects the algorithm with the smallest estimated risk.

While the data-splitting heuristic is very general, a key assumption of CV is that the data is independently and identically distributed (IID). In the following sections, we will see that, for time series data, this is often not the case and requires a different approach.