Data Science · Chapter 29 of 43

Linear Regression

LINEAR REGRESSION fits a straight line (or hyperplane) to predict a numeric target.

Simple, fast, interpretable — a great baseline for regression problems.

Example 1 (python)

from sklearn.linear_model import LinearRegression
import numpy as np
X = np.array([[1],[2],[3],[4]])
y = np.array([2,4,6,8])
m = LinearRegression().fit(X, y)
print(m.predict([[5]]))

Output

[10.]

Learns y = 2x.

Example 2 (python)

print('slope', m.coef_[0], 'intercept', m.intercept_)

Output

slope 2.0 intercept 0.0

Model coefficients.

Key points

Predicts continuous values.
Fits by minimising squared error.
Fast and interpretable.
Assumes an (approx.) linear relationship.

💡 Note: Always look at residuals — patterns in residuals mean linear regression is the wrong model.

Linear Regression

Key points

📝 Quick Quiz