ST 730  TIME SERIES                                            
FALL 2004

*-----------------------------------------------------------*
| D. A. DICKEY, 17 Patterson Hall                           |
|     office hours 9-10 Monday, 9-10 Wed.or by appt.        |
|     dickey@stat.ncsu.edu, (919)-515-1925                  |
| Grader: Adam Hoppes rahoppes@stat.ncsu.edu                |
|     office hrs.  10-12 Wed.,  9 Patterson Hall            |
|                                                           |
| Course meets in Polk 5-A                                  |
| Tues. & Thurs.  1:05-2:20                                 |
*-----------------------------------------------------------* 
                    
TEXT: 
SAS System for Forecasting Time Series 2nd ed. (Brocklebank & Dickey)
We will jump to chapter 7 early to ensure coverage of frequency domain
      
TENTATIVE SCHEDULE:

Week of

Aug.  19   Basic operations sorting, merging handling dates, plotting   Online notes
Aug.  17   PROC EXPAND, regression, forecasts                               "
Aug.  31   Stationarity,  ACF,  AR Models, MA models                    Chap 2-3
Sept.  7   ARMA, IACF, PACF estimation methods                              "
Sept. 14   ESACF, SCAN table etc.                                           "
Sept. 21   Transformations, unit root tests, diagnostics                    "

                    Test 1   Tues. Sept. 21

Sept.28    Trigonometric Regression and Periodograms                    Chap 7
Oct.  5    Spectrum of ARMA processes                                      "
Oct. 12    Smoothing and interpreting the spectrum                         "
Oct. 19    Linear filters                                                  "
Oct. 26    Regression with AR Errors, PROC AUTOREG, ARCH models         Chap 4,5

                    Test 2   Tues. Oct. 26

Nov.  2    Seasonality, Transfer Functions                              Chap 4
Nov.  9    Intervention Models                                          Chap 4
Nov. 16    VAR and Cointegration                                        Chap 4,5
Nov. 23    Statespace models                                            Chap 6

                    Test 3   Tues. Nov. 18   

Nov. 30    Bivariate Spectral quantities                                Chap 7

Final Exam is scheduled for 
   Thurs. Dec. 9
   1 to 4 pm  (in our classroom)


TESTS:  
Tests and the percentages they count are:

    Test 1:   20% of grade
    Test 2:   20% of grade
    Test 3:   20% of grade
    Final:    25% of grade

For each test you can bring with you 1 sheet of paper (standard size)
with anything you like written on it (both sides).  Other than that,
the tests are closed book.  For the final you can bring 3 pages.

HOMEWORK will be assigned at the discretion of the instructor and will 
count the remaining 15% of your grade. Do not start working homeworks 
until due dates appear.  

FINAL: 
On the last day of class I will estimate a final exam grade for you, 
based on your tests and homework thus far, and I will announce the 
cutoff points for letter grades. If you are happy with your grade
based on the estimated final, you are finished.  If you opt to take 
the final exam, I will replace the estimated exam by the actual final 
exam grade if the actual final exam grade is higher. 

INTEGRITY: 
University regulations require that every course syllabus
remind students that the Code of Student Conduct defines a university
policy on academic integrity already pledged by each student. Instructors
assume that the students' names on their submitted work imply compliance
with this policy.  See
    
             http://www.fis.ncsu.edu/ncsulegal/41.03-codeof.htm 

some specifics- 
For HOMEWORK it is OK to talk to others about how to work problems and
even to compare and discuss answers.  It is definitely not OK to 
copy any portion of someone else's homework by machine or by hand.
That is plagiarism. 

Though it should not need to be mentioned, there is to be no 
communication of any kind - verbally, electronically or visually 
with any other student during any TEST or EXAM.