How to choose the Dataset for Retraining the project?

[xda1]

Hi, i am a new user to SAS Viya. When I am trying to use SAS Viya to manage new models (such as credit scoring), I am confused about what are the ways to select retraining datasets. For example, when evaluating the performances on the tested dataset (Q1,Q2,Q3,Q4,Q5) by looking at the ROC graph Q2 and Q5 are doing the best, and there are certain variables showing great differences across the quarters which requires retraining our model on other data. I wonder how do we choose the data we should use to retrain then? What are the standards or ways to select the data we use? It will be great If anyone can help me with this question, thank you!    ( this is the link for the tutorial I am looking at, https://www.youtube.com/watch?v=IGW77r-KQDc , in their example, we will use the most recent data Q5’s data to retrain as the model is degrading over time, but if the degrading is not following a timely order? )