Spring Quarter 2016

Business 41202: Analysis of Financial Time Series

Instructor: Ruey S. Tsay

Phone: 773-702-6750

Fax: 773-702-0458 (Please put my name on the cover
page)

Office HPC: 455

Lecture:

Bus 41202-01: Friday AM to 11:30 AM at Class Room 10, Harper Center

Bus 41202-85: Saturday 1:30 PM to 4:30 PM at NBC 130, NBC Tower

Teaching Assistant: Mr. Jae Hyen Chung

e-mail: jchung4@chicagobooth.edu

(e-mail is the easiest way to contact Jae)

Review Sessions:

BS41202-01: Thursdays 12:15 PM to 1:15 PM (Room HCC09). (Starts from Week 2.)

BS41202-85: Saturdays 12:15 PM to 1:15 PM (Booth 455-130). (Starts from Week 1.)

Syllabus of the course.

Course materials

Text: Analysis of Financial Time Series,
3rd Edition

Ruey S. Tsay,
Wiley, 2010.

ISBN:
0-470-41435-4

Data sets:

http://faculty.chicagogsb.edu/ruey.tsay/teaching/fts3/

or additional datasets will be posted for lectures and homework
assignments.

Lecture Notes:
posted a week before lecture. (Data sets will be updated)

>> Print a copy of the lecture notes before class.
<<

Pre-class reading: Note and data sets: d-ibm-0110.txt, d-vix0411.csv

Week 1: Lecture & Data sets used: d-aapl0413.txt, d-exuseu.txt, m-ibm-6815.txt, m-tb3ms.txt

m-tb6ms.txt, q-ko-earns8309.txt, taq-cat-t-jan042010.txt, d-cdsJPM.txt

Week 2: Lecture & Data sets used: dgnp82.txt, d-ibmvwew6202.txt, m-dec12910-6114.txt

R script: lagplot.R

Week 3: Lecture & Data sets used: q-gdpc96.txt, w-gs1n36299.txt, m-hstarts-5912.txt, q-earn-jnj.txt, m-tb3n6.txt,

UNRATE-1.txt

Week 4: Lecture & Data sets used: VIX and SP5 index downloaded via the quantmod package, m-intc7303.txt, sp500.txt

(start to use the fGarch package).

Week 5: Lecture & Data set used: m-ibm2609.txt

(Addtional packages used: Introduction) rugarch and stochvol. The introduction gives examples.

Week 7: Lecture & Data set used: download directly using quantmod.

Week 7a: Lecture & Data sets used: m-gmsp6708.txt, m-ibmln2699.txt, taq-jnj-t-oct4t152010.txt, ibm91-ads.txt, ibm91-adsx.txt,

taq-cat-t-jan042010.txt, taq-cat-cpch-jan042010.txt

Week 8:Lecture, Data sets used:

Week 9: Lecture & portfolio, Data set: d-ibm-rq.txt, d-ibmbaml3a-0110.txt, d-ibm-0110.txt

R packages used: evir, quantreg, and fGarch.

Week 10: Lecture & Data sets used:q-gdpun.txt, d-bhp0206.txt, d-vale0206.txt, d-ibmbaml3a-0110.txt;

R packages used: urca and MTS

Computing:

The main software used is R, which is free
from R-Project for Statistical Computing.

The most recent version of R is R3.1.2

The following
packages are needed in R:

(fBasics,
fGarch, quantmod, fUnitRoots, MTS,
nnet, evir, stochvol, urgarch, urca)

[In fact, you may want to install the complete package Rmetrics. This can be done in R using

the following two commands:

source("http://www.rmetrics.org/Rmetrics.R")

install.Rmetrics()
]

[* Students may use other packages or programs if they prefer. *]

R Installation:

Instructions
for download R:

1. Download R here. [Click CRAN to select a mirror site; a web site close to you.]

Instructions
for running R on PC will be demonstrated in class.

[* Students may also use RStudio (which can also be found free on the web). *]

R commands used in lectures: (updated weekly)

Week 1: Rcommands_lec1.txt

Week 2: Rcommands_lec2.txt

Week 3: Rcommands_lec3.txt

Week 4: Rcommands_lec4.txt

Week 5: Rcommands_lec5.txt

Week 6:

Week 7: Rcommands_lec7.txt

Week 8: Rcommands_lec8.txt

Week 9: Rcommands_lec9.txt

Week 10: Rcommands_lec10.txt

Homework :

Assignment before class: Read Chapter 1 of the textbook.

HW1: Data sets used: d-sbux3dx-0715.txt, m-aapl3dx-0115.txt, d-exuseu-0516.txt

Solutions & R-commands used. The R output is for your reference only.

HW2 : Data sets used: m-globaltemp.txt, m-COILWTICO.txt

Solutions & R-commands used.

HW3 : Data sets used: m-dec12910-6114.txt, bond6m-6115.txt, d-cdsJPM.txt

Solutions & R-commands used.

HW4 : Data sets used: m-ibmsp-6115.txt

Solutions & R-commands used.

HW5 : Data sets used: m-pgsp-4115.txt, taq-t-sbuxdec2031-2014.txt (Due May 27 and 28, respectivley).

Solutions & R-commands used.

HW6: Date sets used: taq-sbux-pch-dec22-2014.txt, (also use quantmod to download the stock data.)

Solutions & R-commands used.

Office hour
: (a) Thursday: 4:00 PM to 5:00 PM (Harper Center Rm 455)

(b) By appointment

(c) E-mail me at any time with questions.

(This is the easiest way to reach me.)

Midterm :
Week 6. Open book and notes!

Date: Campus session: May 6 (Friday)

Weekend session: May 7 (Saturday)

Final Exam:
Exam week

Campus: Friday, June 10, 8:30 AM to 11:30 AM

Weekend: Saturday, June 11, 12:30 PM to 3:30 PM (Graduating); 1:30 PM to 4:30 PM (other students)

Open book and notes.

Grading:
30% midterm + 35% final exam + 5% in-class discussion + 30% homework,

where the scores of each component are normalized to be out of 100.

Chicagobooth mandates a maximum class
grade point average of 3.33.

I rank the class based on the final scores using the above grading

formula and pick grade cutoffs so that I can get the highest class

GPA under the constraint.

Provisional Grade: For graduating students, provisional grades will be assigned based on the midterm and hw assignments

that are available via Week 8.

Additional
Web Sites for data:

(a) Wharton WRDS at http://wrds.wharton.upenn.edu

(b) St Louis Fed at http://research.stlouisfed.org/fred2/

Additional R scripts:

0. Moving average plot: ma.R

1. Forecasts with specified forecast origin: fore.R & forecast plot: foreplot.R ; one-step predictions: foreOne.R

2. Recursive out-of-sample forecasts: backtest.R

3. To estimate an IGARCH model: Igarch.R & GARCH-M program: garchM.R

4. Moving window for volatility calculation: mvwindow.R

5. Yang and Zhang's methof: yz.R

6. Recursive out-of-sample nnet forecasts: backnnet.R

7. Compute volatility of a given GARCH11 model: garch11v.R

(including IGARCH(1,1) model for RiskMetrics).

8. Co-integration test: (Use urca package)

9. ACD estimation: acd.R

10. High-frequency intraday log returns: hfrtn.R

11. High-frequency intraday number of transactions: hfntra.R

12. High-frequency price change and duration: hfchg.R

13. Compute VaR based on traditional EVT estimates: evtVaR.R

14. Risk measure: RMeasure.R

15. TGARCH11 & exp-GARCH models: Tgarch11.R & Egarch.R

16. R program for SV model: svfit.R (no leverage effect); See also the R package "stochvol"

17. R program for RiskMetrics: RMfit.R

Feedback: file

Final Exam: Exam and Solutions

Old exams :

Year 2014: midterm and solutions

Final exam and solutions

Year 2015: midterm and solutions

Final exam and solutions