In: Statistics and Probability
Can you use linear regression framework to model non-linear relationships? Explain how.
The term linear regression means that the , response variable is linear in the unknown coefficients that is betas ( not in X ) This type of regression equation is linear in the parameters. However, it is possible to model curvature with this type of model. While the function must be linear in the parameters, you can raise an independent variable by an exponent to fit a curve. For example, if you square an independent variable, the model can follow a U-shaped curve.
For example, linear regression to fit quadratic curve, y = b0 + b1*x + b2*x*x, it is still linear regression as it is linear in parameters