ch12sol - Chapter 12 Regression Models 12.1 The point ( , y...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Chapter 12 Regression Models 12.1 The point (ˆ x 0 , ˆ y 0 ) is the closest if it lies on the vertex of the right triangle with vertices ( x 0 ,y 0 ) and ( x 0 ,a + bx 0 ). By the Pythagorean theorem, we must have h x 0 - x 0 ) 2 + ( ˆ y 0 - ( a + bx 0 ) ) 2 i + h x 0 - x 0 ) 2 +(ˆ y 0 - y 0 ) 2 i = ( x 0 - x 0 ) 2 + ( y 0 - ( a + bx 0 )) 2 . Substituting the values of ˆ x 0 and ˆ y 0 from (12.2.7) we obtain for the LHS above " ± b ( y 0 - bx 0 - a ) 1+ b 2 ² 2 + ± b 2 ( y 0 - bx 0 - a ) 1+ b 2 ² 2 # + " ± b ( y 0 - bx 0 - a ) 1+ b 2 ² 2 + ± y 0 - bx - a ) 1+ b 2 ² 2 # = ( y 0 - ( a + bx 0 )) 2 " b 2 + b 4 + b 2 +1 (1+ b 2 ) 2 # = ( y 0 - ( a + bx 0 )) 2 . 12.3 a. Differentiation yields ∂f/∂ξ i = - 2( x i - ξ i ) - 2 λβ [ y i - ( α + βξ i )] set = 0 ξ i (1 + λβ 2 ) = x i - λβ ( y i - α ), which is the required solution. Also, 2 f/∂ξ 2 = 2(1 + λβ 2 ) > 0, so this is a minimum. b. Parts i), ii), and iii) are immediate. For iv) just note that D is Euclidean distance between ( x 1 , λy 1 ) and ( x 2 , λy 2 ), hence satisfies the triangle inequality. 12.5 Differentiate log L , for L in (12 . 2 . 17), to get ∂σ 2 δ log L = - n σ 2 δ + 1 2( σ 2 δ ) 2 λ 1+ ˆ β 2 n X i =1 h y i - α + ˆ βx i ) i 2 . Set this equal to zero and solve for σ 2 δ . The answer is (12 . 2 . 18). 12.7 a. Suppressing the subscript i and the minus sign, the exponent is ( x - ξ ) 2 σ 2 δ + [ y - ( α + βξ )] 2 σ 2 ± = ± σ 2 ± + β 2 σ 2 δ σ 2 ± σ 2 δ ² ( ξ - k ) 2 + [ y - ( α + βx )] 2 σ 2 ± + β 2 σ 2 δ , where k = σ 2 ± x + σ 2 δ β ( y - α ) σ 2 ± + β 2 σ 2 δ . Thus, integrating with respect to ξ eliminates the first term. b. The resulting function must be the joint pdf of X and Y . The double integral is infinite, however. 12.9 a. From the last two equations in (12.2.19), ˆ σ 2 δ = 1 n S xx - ˆ σ 2 ξ = 1 n S xx - 1 n S xy ˆ β , which is positive only if S xx > S xy / ˆ β . Similarly, ˆ σ 2 ± = 1 n S yy - ˆ β 2 ˆ σ 2 ξ = 1 n S yy - ˆ β 2 1 n S xy ˆ β , which is positive only if S yy > ˆ βS xy .
Background image of page 1

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

View Full DocumentRight Arrow Icon
12-2 Solutions Manual for Statistical Inference b. We have from part a), ˆ σ 2 δ > 0 S xx > S xy / ˆ β and ˆ σ 2 ± > 0 S yy > ˆ βS xy . Furthermore, ˆ σ 2 ξ > 0 implies that S xy and ˆ β have the same sign. Thus S xx > | S xy | / | ˆ β | and S yy > | ˆ β || S xy | . Combining yields | S xy | S xx < ± ± ± ˆ β ± ± ± < S yy | S xy | .
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 5

ch12sol - Chapter 12 Regression Models 12.1 The point ( , y...

This preview shows document pages 1 - 3. Sign up to view the full document.

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