The algebraic expression of Linear Regression with multiple predictors can be derived from the n-NN model.

• A. True
• B. False
• C. Not enough information
• D. Both

Answer: (B)

Linear regression with multiple predictors has an explicit algebraic form where the predicted value is a linear combination of the predictors, y = β0 + β1x1 + ... + βp xp, obtained by minimizing the sum of squared residuals and solving the normal equations. The n-NN model (nearest-neighbor) is a non-parametric approach that predicts y by averaging the responses of the closest training points, without assuming a global linear relationship or yielding a single set of coefficients that apply across the entire feature space. Because k-NN relies on local similarities rather than a global parametric equation, there isn’t a standard algebraic expression for linear regression that is derived from it. In some contexts you can approximate a linear function locally with a neighborhood, but that becomes local linear regression and is not the same as the global linear regression model. So the statement is not correct.