STAT2004J – Linear Modelling Tutorial 7
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
STAT2004J – Linear Modelling
Tutorial 7
Question 1
We assume that the random variables Y1, Y2 , Y3 , Y4 are independent with E(Yi) = μi and Var(Yi) = σ2 . We write
a) Write the covariance matrix of Y in matrix form, i.e., Cov(Y) . The covariance matrix of a random vector X = (X1 , · · · , Xn) is then Xn matrix whose (i, j)-entry equals Cov(Xi, Xj).
b) We deine
where W1 = Y1 + Y2 + Y3, W2 = Y1 — Y2, W3 = Y1 — Y2 — Y3. The objective is to calculate Cov(W) in two diferent ways.
i) Calculate manually Cov(Wi, Wj) for any i, j = 1, 2, 3 and form Cov(W).
ii) Show that there is a deterministic matrix A such that W = AY and calculate Cov(W) using the fact that Cov(AY) = A Cov(Y)AT .
Question 2
Let
f(x) = 3x1(2) + x2(2) — x1x2 + 2x1 — 5x2 + 1.
a) Let x = [x1 , x2]T . Find A, b and c such that
f(x) = xTAx + bTx + c,
where A is a symmetric matrix.
b) Solve for = 0.
Question 3
The observations y1 , y2 and y3 were taken on the random variables Y1, Y2 and Y3, where Yi’s satisfy the following equations with unknown deterministic constants θ and φ:
Y1 = θ + E1 Y2 = 2θ — φ + E2 Y3 = θ + 2φ + E3
E(Ei) = 0, Var(Ei) = σ2 (for i = 1, 2, 3), Cov(Ei, Ej) = 0 (i j)
Formulate this model in matrix notation, and hence ind the vector of least squares estimates of the regression parameter vector . Also calculate the covariance matrix of the least
square estimator of β .
2024-01-15