Statistics 679: Time Series Analysis II
- Course information, assessment procedures,
notes. These are constantly under revision, before and after class.
- The course assumes an
undergraduate knowledge of Time Series at the level of STAT 479,
and mathematics at the level of STAT 512.
and using the R package:
- For a quick start you
can obtain R here and
install it on a PC by double-clicking on this .exe file and following the instructions.
There is a Mac version of R too; you'll need to consult the 'R home page'
link below for this.
- There is a problem with
R version 2.8 which impacts us - see the notes by
S&S. But version 2.7 is fine and can be obtained here.
- What is R?
- The R home page
- An R manual
brief introduction to R
- Typographical errors in
- Some further (and very
useful) information about
R; see in particular the authors' R
tutorial and some of their caveats
regarding R for time series analysis.
- Datasets used in the text.
These can be saved to your own system individually, or you can download
all of them at once as data.zip.
- R code used in
- Some R functions supplied by Shumway
& Stoffer and extended and modified by me. See the notes by S&S.
You can make their (unmodified) functions available by entering a command source("http://www.stat.pitt.edu/stoffer/tsa2/Rcode/itall.R") within
any R programme.
- Further R
functions and code
unique to Chapter 6. You can make these functions available by
entering a command source("http://www.stat.pitt.edu/stoffer/tsa2/Rcode/New6/ssall.R") within
an R programme.
- Code unique to
- The importance of
clarity of exposition, and of grammatical correctness, in technical
writing cannot be over-emphasized. The best way to determine if you
understand what you are doing is to try to write it down in a form that
another reader can understand. Some comments along these lines are
included with the course information; for
some helpful resources see "Writing Aids" on my homepage.
for R examples discussed in class (some of these originated at the authors' website):
You should find it useful to download and run these R scripts before
the classes in which they are to be discussed.
- Lecture 1 - SOI and
Recruits; ACF, CCF, regression.
- Lecture 2
- Varve series; estimation and forecasting (Example 3.31).
- Lecture 4a - SOI and
Recruits; spectrum estimation, coherence (Examples 4.11, 4.16).
- Lecture 4b - SOI
and Recruits; impulse-response; optimal filtering
(Examples 4.23, 4.24).
- Lecture 5 - Varve; long
memory (Example 5.1).
- Lecture 6a - NYSE; GARCH fit
- Lecture 6b - Flu; threshold
modelling (Example 5.5).
- Lecture 7a - Mortality;
regression modelling (Example 5.6).
- Lecture 7b - SOI and
Recruits; transfer functions (Example 5.8).
- Lecture 9 - VAR modelling of Mortality
data (Examples 5.9, 5.10, 5.11).
- Lecture 12 - Kalman filter and
smoother on simulated data (Example 6.5).
- Lecture 13a - Global Temperature
series; ML estimation (Example 6.7).
- Lecture 13b - Simulated data;
ML estimation via EM algorithm (Example 6.8).
- Lecture 14 - J&J
Quarterly Earnings; structural modelling (Example 6.10).
- Lecture 15 - Newbold data;
bootstrapping (Example 6.12).
19 - Shasta Lake data; multiple regression in frequency domain
(Example 7.1); calls stoch.reg.R.
- Lecture 20 - "Nuke"
data; regression with deterministic inputs (Example 7.4).
Contributed by Fraser Newton.
- Lecture 21 -
"Nuke" data; regression with random coefficients (Example
7.6). Contributed by Junfeng Ma.
- Lecture 22a - fMRI data (all
234 series) arranged so as to produce Figure 22.1.
- Lecture 22b - fMRI anova (Examples
7.7, 7.8). Contributed by James Ng.
- Lecture 23 - Principal component
analysis of fMRI data (Example 7.14).
- Lectures 24a
- Discriminating earthquakes from explosions in time domain (Example 7.11).
- Lecture 24b
- Discriminating earthquakes from explosions in frequency domain (Example
- Lecture 25 - Clustering
earthquakes and explosions (Example 7.13).
Statistics resources at U of A:
of Doug Wiens
of Robert Shumway
of David Stoffer
Department of Mathematical and Statistical
Sciences Home Page