An Introduction to Time Series Analysis and Forecasting with Stata

TBA
Cass Business School, 106 Bunhill Row, London EC1Y 8TZ, UK


Course Description
Course Programme
Request an Enrolment Form Now
Terms and Conditions
Stata Software

Timberlake Consultants Ltd, the UK distributor of Stata, invite you to attend a four days course covering the use of Time Series Analyses with Stata 10, the well known statistical software package software package developed by StataCorp (USA).

The Course - This course provides an introduction to Time Series Analysis and Forecasting with Stata. The course assumes little mathematical background on the part of the participants. It teaches theory, modeling, programming, and interpretation of the major time series models, along with interesting applications to business and risk analysis in finance on a Windows based platform. The course shows how to apply these techniques to real-life social science, economic, business, financial, and medical data, with many examples on the reporting and interpreting of the results. Participants are welcome to bring their own data.

Who should attend - The course, given in English, is aimed at students, researchers, and forecasters interested in

  • Longitudinal analysis with Stata
  • Box-Jenkins Time Series Analysis with Stata
  • Seasonal Box-Jenkins Models
  • Outlier modeling
  • Dynamic Regression (impulse response function) Analysis with Stata
  • GARCH modeling with Stata
  • Forecasting with time series models
  • Forecasting evaluation
  • Policy and Impact Analysis with Stata
  • Financial Risk Analysis with Stata

Mathematical Background Required

  • High School Algebra
  • Basic Statistics

Helpful but not required background 

  • Linear or Matrix Algebra
  • Basic differential and integral calculus

Advantages - The course will

  • Provide an Introduction to Applied Time Series Analysis Theory, Modeling, and Forecasting with Stata
  • Review major Time Series Analysis and Forecasting Theory including Box-Jenkins ARIMA, Time Series Regression, and GARCH Modeling
  • Provide hands-on experience in time series analysis and forecasting models - each delegate is provided with a computer throughout the course

The Principal Lecturer – Dr Robert A. Yaffee.

           Robert A. Yaffee, Ph.D., is currently a research professor at New York University, and from 1989 until spring 2004 served as a senior research/statistical consultant with New York University’s Academic Computing Services.  Dr. Yaffee is an author of a forthcoming textbook entitled An Introduction to Forecasting Time Series using Stata (expected publication date Winter 2006). He is also an author of a recent textbook An Introduction to Time Series Analysis and Forecasting with Applications of SAS and SPSS (Academic Press, 2000).

Cost - The cost of the course is:

Organization Type
1st Participant
2+ Participants
Commercial
£1,600+VAT=£1,880.00
£1,450+VAT=£1,703.75
Academic
£1,100+VAT=£1,292.50
£1,100+VAT=£1,292.50

The cost includes course materials, all lunch and refreshments and the use of computers. The number of delegates is restricted. Please register early to guarantee your place. Further instructions will be sent with the joining instructions. If you need assistance in locating hotel accommodation in the area, request the help of our Training Department.

Payment can be made by cheque, bank transfer or credit/debit card (a surcharge of 2% applies to credit card payments only. There is no charge for debit card payments).

Day 1   Morning
8.30am coffee and Registration

9.00am
1.Basic Time Series Analysis Concepts

  • definition of a time series
  • cycles
  • trends
  • seasonality
  • lags, leads, differences
  • nomenclature
  • Expectation notation
  • Summation notation

10:30am Break
2. Time Series Setup with Stata

  • inputting time series data
  • time-date functions and applications
  • importing and exporting time series data
  • graphing Time Series with Stata
  • preliminary analysis of time series with Stata

12:00 noon - 1:30 pm Lunch
Day 1  Afternoon 1:30

3. Stationarity

  • covariance stationarity
  • strict stationarity
  • Dickey Fuller tests: theory & programming dfuller tests
  • Augmented Dickey-Fuller tests: Theory & programming
  • Phillips-Perron tests

4. Autocorrelation

  • Theory
  • Types
  • Characteristic ACF and PACF patterns
  • Programming the correlogram
  • Box-Ljung significance tests

2:30-2:45pm Break
5. Moving averages

  • Theory
  • Types
  • Characteristic ACF and PACF patterns
  • Programing the ACF and PACF
  • White noise Significance tests

4:00pm
6. Hands-On  Experience and Programming practice 

  • Stationarity diagnosis and transformations
  • ARIMA identification
  • Integrated processes    
  • AR processes
  • MA processes
  • ARMA processes

Day 2  - Session begins at 9:00am

1. ARIMA modeling

  • estimation
  • estimation algorithms
    • full maximum likelihood
    • conditional maximum likelihood
  • diagnosis
  • Intervation modeling
  • model fitting

10:30am Break
2. Seasonal ARIMA models

  • Identification
  • Estimation
  • diagnosis
  • model fitting

12:00 noon - 1:30pm Lunch
Day 2:  Afternoon 1:30- 2:30pm

  • Forecasting theory
  • sample segmentation
  • segment lengths
  • in-sample v. post-sample forecasting
  • point forecasts
  • interval forecasts
  • forecast profiles
  • out-of-sample forecasts
  • one-step forecasts
  • dynamic forecasts
  • structural forecasts
  • combining forecasts

2:30-2:45pm Break
3. Forecasting Evaluation

  • Test of forecast bias
  • Test of forecast accuracy: out-of-sample and ex-ante
  • MSFE
  • MAE
  • MAPE
  • MdAPE
  • Theil's U
  • Diebold-Mariano test of comparative forecast evaluation

4.Forecasting Graphics

3:00-5:50pm
5. Hands  on  ARIMA modeling and forecasting

Day 3 Session begins at 9:00am

1. Intervention (Impact) Analysis

  • Pulse interventions
  • Level Shifts
  • Testing for them

2. Outliers

  • Additive
  • Seasonal Pulses
  • Innovational
  • Patches
  • Modeling outliers

3. Intervention modeling with Arimacheck

4. Hands-on programming

12:00 noon -1:30pm Lunch
Day 3:  Afternoon    

Transfer Function/Impulse Response Function Analysis

  • Dynamic Regression Models with Stata
  • Impulse Response functions
  • Deterministic inputs
  • Stochastic inputs
  • Dynamic Regression Analysis Linear Transfer Function methodology
  • Dynamic Regression modeling with airmacheck
  • Forecasting with Dynamic Regression Models
  • out-of-sample
  • ex ante

2:30-2:45pm Break

  • Cointegration
  • Exogeneity
  • Granger causality
  • Tests for Exogeneity
  • Error Correction models

4:00-5:00pm

Q and A

Hands on programming

Day 4:  Session begins at 9.00am

Autoregressive Error Models

  • First order correction theory: Cochran-Orcutt
    • Prais-winston models
    • Newey-west robust models
    • Regression diagnostics
      • autocorrelation tests
      • heteroskedasticity tests
      • parameter constancy tests

10:30-10:45am Break

Q and A

Hands-on programming

Robust time series analysis

  • Semi-robust time series analysis
  • Robust time series models
  • Robust time series with arimacheck

12:00- 1:30pm Lunch

GARCH models: Theory and programming

  • ARCH
  • GARCH
  • IGARCH
  • EGARCH
  • GARCH-in-Mean
  • Forecasting Volatility with GARCH
  • Volatility smiles and skews
  • graphing
  • modeling
  • Forecast Evaluation with GARCH Forecasts
  • out-of-sample
  • ex ante

2:30-2:45pm Break

Recapitulation

Q and A

Hands-on  programming  

5:00pm - End


Request a Registration Form now
Top   

Terms and Conditions

Registration closes 3 calendar days prior to the start of the course. After registering, you are liable for the payment of the fee unless you cancel your registration. Cancellation rules are listed below.

Cancellations:

  • full fee returned for cancellations made over 28 calendar days prior to start of the course
  • half-fee returned for cancellations made 14 calendar days prior to he start of the course
  • no fee returned for cancellations made less than 14 calendar days prior to the start of the course.

Back to Stata homepage
Back to Timberlake Consultants

©Timberlake Consultants Limited
Last revised:20/09/2008