Scribd
Upload
46 views

Time Series Analysis Intro

This document outlines the syllabus for a time series analysis course. It covers topics like autoregressive, moving average, and ARIMA processes. It details the course schedule by week, textbook information, instructor details, homework and grading policies.

Uploaded by

amitag007
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
46 views

Time Series Analysis Intro

This document outlines the syllabus for a time series analysis course. It covers topics like autoregressive, moving average, and ARIMA processes. It details the course schedule by week, textbook information, instructor details, homework and grading policies.

Uploaded by

amitag007
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 13

Stat 131

Time Series Analysis:


Introductory Examples
Instructor: Jose Blanchet (Science center
606)
Instructors office hours (tentative): Mon
2:15 3:15 PM & Wed 5:45 6:45 PM
TF: Wei Zhang
Office hours: TBA
www.courses.fas.harvard.edu/~stat131/
Primary text: Brockwell, P., and Davis, R.
(1991) Time Series: Theory and Methods.
2nd edition, Springer.
Secondary: Brockwell, P. and Davis, R.
(2002) Introduction to Time Series and
Forecasting. 2nd edition. Springer.
Description:
Model building, data analysis, inference and
forecasting using Auto regressive (AR),
moving average (MA), ARMA, and ARIMA
processes.
Stationary and non-stationary processes,
seasonal processes, auto-correlation and
partial auto-correlation functions,
identification of models, estimation of
parameters, and spectral analysis.
Prerequisite: Stat 110, recommended Stat
111 and/or Stat 139.
Weeks Topics Chapters
Background, examples, objectives,
1-2 1.1,1.2,1.4,1.6
general approaches.
Stationary processes: sample mean
3-4 1.3, 1.5, 7.1, 7.2
and autocorrelation.
4-6 Best linear mean square prediction 2.1-2.7
6-7 ARMA processes: autocorrelation 3.1-3.5
Recursive prediction of ARMA
7-8 5.1-5.5
processes.
Parameter estimation for ARMA
9-11 8.1-8.9
processes.
12-13 Model-building with ARIMA 9.1-9.6
13-14* Spectral analysis and filtering 4.1-4.4
Grading Policy
Homework 20% (About 6 problems sets)
May discuss homework problems with other students,
but must write them up independently.
Correct answers without supporting work will not receive
credit.
Homework is due at the beginning of class on the due
date. Late homework will not be accepted.
Your lowest homework score will be dropped when
computing your final grade (i.e. this leaves room for
emergencies!)
SOME HOMEWORKS WILL REQUIRE USE OF
MATLAB OR ANY OTHER COMPUTER PACKAGE
Examples of Time Series
S eries

3000.

2500.

2000.

1500.

1000.

500.

0 20 40 60 80 100 120 140

Monthly sales of red wine (in liters) of an Australian


wine maker MONTLY from January 1980 to October 1991
Examples of Time Series
S eries

3000.

2500.

2000.

1500.

1000.

500.

0 20 40 60 80 100 120 140

Monthly sales of red wine (in liters) of an Australian


wine maker MONTLY from January 1980 to October 1991
Examples of Time Series
S eries

11000.

10000.

9000.

8000.

7000.

0 10 20 30 40 50 60 70

Monthly accidental deaths data 1973 to 1978 measured


in thousands of dollars (seasonal components)
Examples of Time Series
Mean reversion of heights of individuals across
generations (Galton and Pearson)
Additional important applications:
Financial time series
Economic variables such as
inflation, exchange rates
Engineering: linear and non-linear
time series, queueing theory, signal
processing
General objective:
Given a random process identify
trend, seasonality and noise
(randomness)
Estimate the structure of the
standardize (without trend and
seasonality) series
Analysis and forecasting

You might also like