DMtutorial2 - bx i + ε i how does the estimated linear...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Tutorial 3 1. Suppose that X follows a uniform distribution, then for the inner points, the local linear kernel estimator and local constant (NW) estimator have the same asymptotic MSE. 2. For model Y = sin(2 πX ) + ε where X uniform (0 , 1) and ε is independent of X . If we apply local linear kernel estimation to estimate the curve, which point are expected to have the biggest absolute bias? 3. The uniform kernel function is defined as K ( v ) = ± 1 , if | v | < = 1 , 0 , if | v | > 1 . Suppose we have the situation of equal spaced design ( x i ,y i ) with x i = i n , which satisfies y i = m ( x i ) + ε i , where ε i ,i = 1 , ··· ,n are IID N (0 , 1). Using NW kernel smoothing to estimate m ( x ) , 0 < x < 1 with h = n - 0 . 2 , what is the bias and variance of the estimator? what are their limits when n → ∞ ? 4. Consider samples ( x i ,y i ) ,i = 1 , 2 , ··· ,n . when a single observation is distorted by a very large value, that is ( x k ,y k ) ( x k ,y k + c ) for a very large c for a fixed k . how does the estimated curve change under such a distortion? If you fit a liner regression model to the data y i = a +
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: bx i + ε i how does the estimated linear regression model change with such a distortion? 5. Another application of the confidence band is to check whether a pre-specified func-tional shape is acceptable. If the specified function is within the band, we accept the functional form. Otherwise, we reject the functional form. Based on this idea, check whether the relation between Y and X in data 1 and data 2 are linear regression 1 models? [Hint: fit a linear regression to the data. check whether the fitted regression is in the 95% pointwise confidence bands] 6. Consider samples ( x i ,y i ) ,i = 1 , 2 , ··· ,n and linear regression model y i = a + bx i + ε i where ε i are IID N (0 ,σ 2 ). Suppose the least squares estimation is used. For any new point x , construct a 95% confidence interval for the regression function m ( x ) = a + bx . 2...
View Full Document

This note was uploaded on 10/04/2010 for the course STAT ST4240 taught by Professor Xiayingcun during the Fall '09 term at National University of Singapore.

Page1 / 2

DMtutorial2 - bx i + ε i how does the estimated linear...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online