Tasks · Foundational · Beginner

Regression

Predicting a continuous numeric value from input features.

What Regression is

Where classification predicts a category, regression predicts a quantity: a price, a duration, a temperature, an expected revenue.

How it works

Models minimise an error measure such as mean squared error or mean absolute error. Residual analysis matters as much as the headline metric, because systematic error patterns reveal missing features.

Why it matters

Forecasting and pricing decisions run on regression, and the same machinery underlies the loss functions used in far larger models.

Common uses

→Sales and demand forecasting
→Property valuation
→Delivery time estimation
→Energy load prediction

Strengths

✓Interpretable in linear form
✓Well-understood diagnostics

Watch for

✓Sensitive to outliers
✓Extrapolates poorly beyond the training range

Continue exploring

Classification Time Series Forecasting Loss Function