Least Squares Regression
A regression method that finds the line y = a + bx minimizing Σ(observed y - predicted y)², where b = Σ(x - x̄)(y - ȳ) / Σ(x - x̄)² and a = ȳ - bx̄.
Real World
Zoopla uses least squares regression on UK housing data to build their automated property valuation model, minimising the squared difference between predicted and actual sale prices.
Exam Focus
Interpret the gradient in context: say 'for each additional [x unit], y increases by [b]' — a contextual interpretation is required for the final mark.
How well did you know this?