3-day MEAFA Professional Development Workshop on Quantitative Analysis Using Stata, 23-25 July 2008

Workshop Description

The workshop is intended for social science researchers who wish to develop quantitative skills for managing and analysing large-scale datasets, understand the theory of time series and panel data research methods and their applications using real data, and learn Stata at the same time.

While the workshop assumes zero knowledge of Stata and basic understanding of statistics and econometrics, it follows a steep learning curve covering almost the entire range of Stata?s data management capabilities and most significant routines for time series analysis and panel data analysis. This is a hands-on workshop, but theory will be thoroughly discussed before applying any routine.

Stata 10

Stata 10 is a complete, integrated statistical package for data management, statistical analysis, graphing and econometric estimation. Stata is fast, accurate and easy to use. For more information visit StataCorp's website.

Computing Facilities

The workshop will take place at the Faculty of Economics and Business computer labs at The University of Sydney. You do not need to bring your own laptop. Stata 10 licenses for Microsoft Windows will be provided.

Which day to attend?

The workshop spans over three days and you may attend any one, any two or all three days (fees vary on the number of days attended). Day 1 introduces Stata 10 and uses the software?s routines to cover a number of data management and data analysis techniques. Day 2 details the theory for modelling univariate time series, forecasting and bivariate causality relationships, and offers extensive applications using Stata. Day 3 details the theory for working with panel data structures and modelling static and dynamic features of panel data, and again offers extensive applications using Stata. If you have no or little experience with Stata then you should attend Day 1 before moving to the rest of the days.

Enrolment

Numbers are limited and places are reserved on a first-come first-served basis. Successful attendees will be notified shortly after reservation and invoices will be issued accordingly. MEAFA maintains a no refund policy following payment. For more information contact meafa@econ.usyd.edu.au

Fees

You may choose which days to attend and fees vary on the number of days: $550 for attending any one day, $825 for attending any two days, and $1100 for attending all three days (prices include GST). There is a 50% discount for a restricted number of places for non full-time employed PhD and honours students.

Fees include extensive course material, data sets, lectures, use of computing facilities, temporary Stata 10 licenses, full catering and opportunity to network with fellow researchers.

Venue

Economics and Business Building H69, The University of Sydney, Ground Floor, Computer Labs, cnr Codrington & Rose Streets, NSW 2006 (see interactive map).

Presenters

Day 1: Wednesday 23 July 2008, Introduction to Stata and Data Management

08:40

Welcoming tea and coffee

09:00-10:40

Introduction to Stata and Data Formats

Stata environemnt; Configuration and features; updates; personalised system; directory management; obtain help and perform search; online sources; Stata syntax; data formats; import, export, load and save datasets; create pseudorandom datasets; review and document the dataset; ordering of dataset; display format

10:30-10:45

Morning break

10:45-12:15

Data Structures and Types of Variables

Categorical data; continuous data; append and merge other datasets; reorganise datasets; numerical, string and date/time variables; manage missing data; create dummy variables; other special purpose variables; convert datasets into summaries

12:15-13:15

Lunch

13:15-14:45

Data Management

Identify usable dataset; validate claims on data structure; identify duplicate observations; alarming and nonsensical data; benefits of filtering

14:45-15:00

Afternoon break

15:00-16:30

Output Management

Logs for output; use of translators for exporting Stata files and output; copy and paste from Stata to text editors and spreadsheets; reproduction of past work; stored results; saved results; the command display; explicit subscripting; important prefixes

16:30-17:00

Questions and User-Specific Issues

Day 2: Thursday 24 July 2008, Time-Series Methods and Applications

08:40

Welcoming tea and coffee

09:00-10:30

Introduction to Time Series Analysis

Time series data structure; time-series frequencies; describing time series; graphing time series; timeline plots; range plot with lines and other useful time series graphs; applications using Stata

10:30-10:45

Morning break

10:45-12:15

Smoothing and Basic Stationary Processes

Time-series smoothing; moving averages; exponential smoothing; nonlinear smoothing; stochastic difference equation models; stationarity; autoregressive AR series; moving average MA series; random walks and white noise; applications using Stata

12:15-13:15

Lunch

13:15-14:45

Complex Stationary Processes and Causality Relationship

ARMA and ARIMA specifications; ARCH models; rationale of causality; Granger causality tests; Granger-Sims causality tests; seminal papers that have applied the causality idea; applications using Stata

14:45-15:00

Afternoon break

15:00-16:30

Model Selection and Forecasting

Sample autocorrelation function; partial autocorrelation function; regression diagnostics; Box-Jenkins model selection; lag order and lag exclusion selection statistics; tests for white noise; Portmanteau test; Dickey?Fuller unit roots test; augmented Dickey?Fuller test; applications using Stata

16:30-17:00

Questions and User-Specific Issues

Day 3: Friday 25 July 2008, Panel Data Methods and Applications

08:40

Welcoming tea and coffee

09:00-10:30

Introduction to Panel Data Analysis

Advantages and disadvantages over cross-sectional, time-series and pooled methods of analysis; some panel data sets; balanced vs. unbalanced panels; panel data dimensions and frequencies; properties of estimators; unbiasedness; efficiency; consistency; applications using Stata

10:30-10:45

Morning break

10:45-12:15

Static Linear Models

Formulation and estimation; one-way and two-way error components; fixed and random effects; the Least Squares Dummy Variable model; the Within, Between and GLS estimators; Hausman test; describing panel data; variance decomposition; graphing panel data; applications using Stata

12:15-13:15

Lunch

13:15-14:45

Dynamic Linear Models

Nickell biases; Anderson-Hsiao IV estimation; the problem of weak instruments; the Generalised Method of Moments; testing for overidentifying restrictions; applications using Stata

14:45-15:00

Afternoon break

15:00-16:30

Issues Specific to Panel Data Analysis

Testing and modelling for cross sectional dependence; sample selection, attrition and stratified sampling; applications using Stata; robust estimation, iid errors vs. AR(1) errors, cluster-robust errors; applications using Stata

16:30-17:00

Questions and User-Specific Issues