You are on page 1of 1

STAT153 – SECTION 5 EXERCISES

Problem 1. (Adapted from Shumway Exercise 3.11)


Let Xt = Wt + θWt−1 where Wt is white noise with variance σw 2.
t
(a) Find the one step ahead forecast Xt+1 = P(Xt+1 |X1 , X2 , . . . Xt ).
(b) Show that the MSE of your estimate is σ 2 (1 + θ2+2n ).
Problem 2. (Shumway Exercise 3.20). Repeat the following numerical exercise three
times. Generate n = 500 observations from the ARMA model given by
Xt = 0.9Xt−1 + Wt − 0.9Wt−1
with white noise iid N (0, 1).
(a) Plot the simulated data and compute the ACF
(b) Next compute and plot PACF.
(c) Fit an ARMA(1,1) model to the data and comment on its similarity to the true
model.
Problem 3. (Shumway Exercise 3.18) Fit an AR(2) model to the cardiovascular mortality
series (cmort) in astsa package using linear regression and using Yule-Walker. Comment
on the differences.

You might also like