The process of checking a model’s performance on new, unseen data. This is often done by reserving part of historical data as a validation (or “test”) set, or by cross-validation. In the forex domain, validation ensures that a trading model will generalize to future market conditions. For example, after training a neural network on one period of forex data, the bot’s predictions are evaluated on a later period to confirm reliability. Metrics (like predictive accuracy or profit on validation trades) help determine if the model is ready for live use or needs retraining/tuning. Proper validation guards against deploying strategies that only worked on past noise.

Back to Robots Glossary Index