The model
Linear regression predicts a continuous target as a weighted sum of features plus a bias. For one feature the prediction is a straight line, y equals w times x plus b. With many features it becomes a flat plane in higher dimensions.
What the weights mean
- Each weight says how much the prediction moves when its feature rises by one unit, holding the others fixed.
- The bias is the prediction when every feature is zero.
- Large absolute weights flag features the model leans on heavily.
Assumptions worth knowing
- The relationship between features and target is roughly linear.
- Errors are independent and have similar spread across the range.
- Strongly correlated features make individual weights unstable and hard to read.
Why it stays popular
Linear regression is fast to fit, easy to interpret, and a strong baseline. If a fancy model cannot beat a clean linear fit, the extra complexity is rarely worth it.
Key idea
Linear regression models the target as a weighted sum of features. Its weights are directly interpretable, making it both a baseline and an explanation tool.