STAT4290 April 15th 2011 Notes

STAT4290 April 15th 2011 Notes - The bootstrap is a general...

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

View Full Document Right Arrow Icon
θ , = - = ( *- ) Varθ 1nn 1i 1nn θi θ ^2 The resamples can also be used to get an empirical distribution for theta-hat -- this lead also to relatively simple hypothesis tests and confidence intervals. Problem: the total number of possible resample, n^n, gets prohibitively lard vary quickly. In practice – taking a sample of the resamples in good enough. Thus, for a sample (X1, X2…Xn) and an estimator theta-hat, select B resamples, or bootstrap samples, and calculate e.g. Nonparametric Bootstrap! = - = ( *- ) Varθ 1B 1i 1nn θi θ ^2 Parametric bootstrap – we have a sample (X1, X2…Xn) assumed to come from a distribution F(x| theta) wuih parameter the theta. We estimate theta with thetahat. Draw X1-star, X2-star from F(x| theta) for each of the B bootstrap samples. How big should B be? Usually, B=500 or 1000 is sufficient. Goodness to Fit Test
Background image of page 1

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

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

Page1 / 2

STAT4290 April 15th 2011 Notes - The bootstrap is a general...

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