Time Series Analysis
Spring 2015
Midterm Exam
Date: May 27, 2015
Kaiji Motegi
Waseda University
Problem-1: Consider ARMA(1, 1):
yt = a0 + a1 yt1 + t + b1 t1 ,
where t is a white noise with variance 2 . Assume |a1 | < 1 so that cfw_yt is covariance
statio
Avd. Matematisk statistik
EXAM FOR SF2943/SF2945 TIME SERIES ANALYSIS/TIDSSERIEANALYS
THURSDAY 31 MAY 2012, 14.0019.00 HRS.
Examiner : Tobias Ryden, tel. 790 8469
Allowed aids: Formulas and survey, Time series analysis (without notes!). Pocket calculator.
Chapter 1 Part I solutions
1.1 Stationarity requires regularity in the mean and autocorrelation functions, so that these quantities (at least) may be
estimated by averaging.
1.2 Code for (a)-(c)
w = rnorm(150,0,1) # 50 extra to avoid startup problems
xa =
Package astsa
February 19, 2015
Type Package
Title Applied Statistical Time Series Analysis
Version 1.3
Date 2014-11-01
Author David Stoffer
Maintainer David Stoffer <stoffer@pitt.edu>
Description Data sets and scripts for Time Series Analysis and Its App
Chapter 2 solutions to problems: 2.1, 2.2, 2.3, 2.6, 2.8
> AIC(fit)/length(cmort) - log(2*pi)
[1] 4.641492
> BIC(fit)/length(cmort) - log(2*pi)
[1] 4.699677
These values are smaller than the values in Table 2.2, so the addition of the lagged
particulate i