![https intime com login https intime com login](https://www.onlineev.com/wp-content/uploads/2021/05/1620568866_maxresdefault.jpg)
To make things intuitive, here is an image for same: The same forecasted data points are then included as part of the next training dataset and subsequent data points are forecasted. Start with a small subset of data for training purpose, forecast for the later data points and then checking the accuracy for the forecasted data points.
![https intime com login https intime com login](https://raw.githubusercontent.com/saumier/GLAMhack2020-Culture-inTime/master/images/Spotlight.png)
The method that can be used for cross-validating the time-series model is cross-validation on a rolling basis. There is a temporal dependency between observations, and we must preserve that relation during testing. In simple word we want to avoid future-looking when we train our model. We cannot choose random samples and assign them to either the test set or the train set because it makes no sense to use the values from the future to forecast values in the past. In the case of time series, the cross-validation is not trivial. Why can’t we use this process in Time Series: Compute the average of the 10 accuracies which is the final reliable number telling us how the model is performing.Repeat step three 10 times to get 10 accuracy measures on 10 different and separate folds.Let’s say you got 10 folds train on 9 of them and test on the 10th.Focus on train set and split it again randomly in chunks (called folds).Split randomly data in train and test set.