Multiple Regression

Multiple regression extends simple regression to several inputs at once, predicting one outcome from many predictors with an equation like y = intercept + b1·x1 + b2·x2 + … . Instead of fitting a line in two dimensions, you fit a flat surface (a plane or hyperplane) through higher-dimensional data.

Its real power is statistical control: by including several variables together, each coefficient estimates that variable's effect while holding the others constant, so you can ask questions like does extra education raise income after accounting for age and experience? The catch is that what you choose to include or leave out shapes the answer, and a left-out confounder can quietly bias every coefficient.