Bootstrap

# 10232012 p kolm 3 the residual bootstrap steps 1

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ECONOMETRICS, VER. 10/23/2012. © P. KOLM. 3 The Residual Bootstrap Steps: ˆˆ ˆ 1. We start by estimating the regression coefficients β0, β1, …, βk , for the linear regression equation y = β0 + β1x 1 + … + βk x k + ε ˆ ˆ using OLS. Then calculate the regression residuals εi ≡ y; − yi for i = 1,..., N ˆ ˆ 2. Randomly draw N “samples” from {ε1,..., εN } , assuming a probability of 1/N ˆ ˆ for each εi , with replacement (thus it is possible to pick the same εi more than once). In this way we obtain a sequence of N “samples.” Call the bootstrapped ˆ∗ ˆ∗ residuals ε1 ,..., εN ˆˆ 3. Generate the bootstrapped dependent variable yi∗ ≡ yi + εi∗ for i = 1,..., N . Now we N have pairs of “bootstrapped samples” ∗ * * * (y1 , x 1,1, x 1,2, …, x 1,k )′ , ..., ∗ * * * (yN , x N ,1, x N ,2, …, x N ,k )′ We can use OLS to calculate our bootstrap parameter ˆˆ ˆ estimates β *, β *, …, β * for the regression equation 0 1 k y * = β0* + β1*x 1*...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online