Question: Suppose the true regression function is linearm(x) x and we estimate the function using the Nadaraya-Watson estimator. Calculate the bias function B(x). Suppose
Suppose the true regression function is linearm(x) Æ ®Å¯x and we estimate the function using the Nadaraya-Watson estimator. Calculate the bias function B(x). Suppose ¯ È 0. For which regions is B(x) È 0 and for which regions is B(x) Ç 0? Now suppose that ¯ Ç 0 and re-answer the question.
Can you intuitively explain why the NW estimator is positively and negatively biased for these regions?
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
