{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Probability and Statistics for Engineering and the Sciences (with CD-ROM and InfoTrac )

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

View Full Document Right Arrow Icon
Stat312: Final Exam Solutions Moo K. Chung [email protected] May 16, 2003 1. We wish to fit paired sample ( x 1 , y 1 ) , · · · , ( x n , y n ) with a linear regression model Y = cx + . We assume that follows a normal distribution with zero mean and variance σ 2 . (a) Estimate c by minimizing the sum of the squared residuals (5pts). (b) Write the log-likelihood function as a func- tion of c and σ (5pts). (c) Find the likelihood estimator of c by differen- tiating the log-likelihood function in (b). De- rive everything (5pts). (d) Either prove or disprove unbiasedness of the estimator you computed in (c) (5pts, no point given if (c) is incorrect). (e) Compute the variance of the estimator you computed in (a) (5pts, no point given if (a) is incorrect). (f) If the sample correlation coefficient of the above paired data is 0 . 5 , what is the sam- ple correlation correlation coefficient of data ( x 1 - ¯ x, y 1 - ¯ y ) , · · · , ( x n - ¯ x, y n - ¯ y ) ? ¯ x and ¯ y are the respective sample means of x i ’s and y i ’s. Prove your result (5pts). Solution. (a) The sum of squared residuals SSE = n i =1 ( y i - cx i ) 2 (2pts). Letting ∂SSE/∂c = 0 , we get n i =1 x i ( y i - cx i ) = 0 . Solving this, ˆ c = n i =1 x i y i / n i =1 x 2 i (3pts). (b) Note that Y i = cx i and Var Y i = σ 2 for some σ . So Y i N ( cx i , σ 2 ) . Then the likelihood function is given by L ( c, σ ) = const.
Background image of page 1

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

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

{[ snackBarMessage ]}