Menu

Train-Test Split

Train-Test Split

Before training the model, the dataset is divided into two parts.

Training Dataset

Used to teach the model patterns present in historical stock prices.

Testing Dataset

Used to evaluate how well the model predicts stock prices it has never seen before.

A common split ratio is:

  • 75% Training Data
  • 25% Testing Data

Separating the data helps evaluate whether the model can generalize well to unseen market data.