Consider the regression model. Y = beta_1 + beta_2 X_2 + beta_3 X_3 + epsilon No

Question

Consider the regression model. Y = beta_1 + beta_2 X_2 + beta_3 X_3 + epsilon Now assume that there are 100 observations and that the available data are in the form of deviation from sample means, so that x_2 = X_2 - X_2, x_3 = X_3 - X_3, and y = Y - Y. Therefore, y = beta_2 x_2 + beta_3 x_3 + epsilon You already know the following results: x'x = [x'_2 x_2 x'_3 x_2 x'_2 x_3 x'_3 x_3] = [10 -5 -5 20], x'y = [15 80], i' Y = 10,000, i' X_2 = 800, and i' X_3 = 1000 And, finally, you also know the inverse of a 2 times 2 matrix: [a c b d]^-1 = 1/ad - bc [d -c -b a] (a) What are the OLS estimates b for beta_2 and beta_3? (b) Since the data are in deviations from the mean, find the intercept (or constant coefficient), beta_1.

Explanation / Answer

a) b^ = (inv( x'x ))*x'y

(inv( x'x )) =

   0.1143    0.0286
    0.0286    0.0571

b^ =
     4
     5

b2 = 4 , b3 = 5

b) b1 = ybar - b2*x2bar - b3 *x3bar

ybar = 100 , x2bar =8 ,x3bar = 10

b1 = 100 - 4*8 - 5*10

= 18

Please rate

Related Questions

Navigate

• Browse (All)
• Browse C
• Subjects

---