This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: e., x n +1 > x n . This implies that as we repeat the iteration, we will get larger value as long as the sequence moves along the range in x n < 1. Likewise, we can find where the sequence moves to for all x n : For x n < 1, f ( x n ) f 00 ( x n ) < 0 thus, x n < x n +1 : x n moves to the right for 1 < x n < 15 / 6, f ( x n ) f 00 ( x n ) > 0 thus, x n > x n +1 : x n moves to the left for 15 / 6 < x n < 4, f ( x n ) f 00 ( x n ) < 0 thus, x n < x n +1 : x n moves to the right for 4 < x n , f ( x n ) f 00 ( x n ) > 0 thus, x n > x n +1 : x n moves to the left When x n = 1 or x n = 4, f ( x n ) f 00 ( x n ) = 0 and x n = x n +1 , and finally the process ends. (d) (Ans) When f ( x ) is continuous and differentiable functions, it has (a) local minimum(s) x min when f ( x min ) = 0 and f 00 ( x min ) > 0, and a local maximum(s) x max when f ( x max ) = 0 and f 00 ( x max ) < 0. Note that in this example we have f (1) = f (4) = 0, and f 00 (1) < 0 and f 00 (4) > 0. We can say, f ( x ) has a local minimum at x = 4 and a local maxmum at x = 1. From the graph, we can see that there is no global minimum of global maximum. 3. (a) (Ans) Since we are interested in the expected change in man’s income, we start by setting F i = 0 . The regression equation becomes Y i = 21 . 3 + 2 . 5 X i + ˆ u i . 3 Therefore the man’s income is expected to change by 2 . 5Δ X i when there is a change in the variable X i . If he acquires an additional year of education (Δ X i = 1), we may expect that his income will rise by $2,500. (b) (Ans) This time we are interested in the expected change in woman’s income, the case when F i = 1. The regression equation becomes Y i = 21 . 3 + 2 . 7 X i 5 . 2 F i + ˆ u i . If she acquires an additional uear of education (Δ X i = 1), her expected income will rise by $2,700. (c) (Ans) First, we notice that X i F i is the relevant variable to the difference in returns to education between gender. Therefore we perform the Ttest to see if the coefficient of X i F i is zero or not. The tstatistic is t = . 2 . 05 = 4 and we find that  t  > 1 . 96 = z . 975 ....
View
Full Document
 Winter '08
 Stohs
 Linear Regression, Regression Analysis, Yi, Errors and residuals in statistics, Xn

Click to edit the document details