Consider the weighted least squares normal equations for the case of simple linear regression where time is
Question:
Consider the weighted least squares normal equations for the case of simple linear regression where time is the predictor variable. Suppose that the variances of the errors are proportional to the index of time such that \(w_{t}=1 / t\). Simplify the normal equations for this situation. Solve for the estimates of the model parameters.
Fantastic news! We've Found the answer you've been seeking!
Step by Step Answer:
Related Book For
Introduction To Linear Regression Analysis
ISBN: 9781119578727
6th Edition
Authors: Douglas C. Montgomery, Elizabeth A. Peck, G. Geoffrey Vining
Question Posted: