Microeconometrics Using Stata
- Course Overview
- Register Now
- Terms & Conditions
Timberlake Consultants, official distributor of Stata, invite you to attend the following 3-day course to be held in Cambridge, UK.
This course provides a review and practical guide to a number of microeconometric models and estimators. We focus on panel and count data models and also examine a broad class of models of discrete choice behaviour. The course emphasise two estimators, the instrumental variable estimator, and the Generalised Method of Moments. The objective of the course is to provide a solid introduction to each of the topics, reviewing both the theoretical econometrics and the empirical literature. Each topic discussed is then illustrated with real data examples using Stata. Some of the examples and applications discussed include: the demand for differentiated products in the ready-to-eat cereal industry, flexible substitution patterns in the new car market, foreclosure in subprime mortgages, the number of patents held by firms, the number of children born to a woman.
The course will be delivered using Stata 12.
Generalised Method of Moments (the link between OLS, method of moments, IV and GMM)
A natural starting point is GMM since this estimator provides a general framework for inference by encompassing a large number of estimators in econometrics. It is not necessary to specify a complete "model" for the process generating the data, and in this regard we observe a more robust estimator. We consider the nature of the generalization in two ways: (i) moments can be nonlinear functions of the unknown parameters; (ii) there may be more moments than unknowns. GMM unifies these two aspects within a single estimation strategy.
We will cover a number of topics including Method of Moments (OLS as MOM), maximum likelihood as a GMM estimator, the optimal weight matrix, and IV as both a MOM and a generalized least squares estimator. We also examine the Generalised IV (GIVE) estimator and nonlinear GMM estimators, with examples from count modelling and discrete choice.
Version 12 of Stata includes a gmm command to compute generalized method of moment (GMM) estimators, making it much easier to code up the nonlinear instrumental variables examples considered in this course.
Discrete Choice Models (conditional and multinomial logit, multinomial probit and mixed logit): Much of the data we collect in economics and the social sciences contains measures of economic activity that are inherently discrete. Common examples are the decision as to which mode of transport to use, which car to buy and whether to work full or part-time.
In this course we also review material from industrial organisation where the choice set is the set of products considered by a consumer. Since in many instances purchase occasions can be thought of as buy at most one, then we see that the discrete choice model is an integral component of many consumer demand systems.
Instrumental Variables Estimation (IV Estimation in linear models, overidentifying restrictions, tests for endogeneity): IV estimation facilitates consistent estimation of model parameters when one or more of the explanatory variables are endogenous. We consider sources of endogeneity that are common in micro-level datasets, including measurement error, omitted variables and simultaneity. The problem of and methods to tackle endogeneity is a rapidly expanding area of research, featured prominently in the widely acclaimed text Mostly Harmless Econometrics (MHE) by Angrist and Pische (2008). We review a number of standard and new methods and consider a number of examples from MHE.
Panel Data (pooled OLS, fixed and random Effects estimators, dynamic panel data methods): Over the last 10 years we have witnessed a steady increase in the availability of datasets that contain observations at the level of the economic agent observed over time. So-called panel data models are now a mainstay of modern econometric methods. In this module we provide an introduction to panel data models commonly used in microeconometric applications, including Dynamic Panel Data Models introduced by Arellano and Bond.
Count Models (the Poisson regression model, extensions of the Poisson model, quasi-maximum likelihood estimation, generalised linear models, mixture models): Count data models are an important component of econometric methodology. Phenomena as diverse as the number of patents held by a pharmaceutical company, the number of visits to a doctor, and the number of children born to a woman, are examples of data that are generated by a count process.
This course is aimed at economists and applied econometricians who deal with different types of data and projects in their day-to-day work. Professionals who are interested to learn different techniques and raise their awareness of possible methodologies that can be used in their current or future projects will greatly benefit from this course.
Basic knowledge of statistics and regression analysis. Previous work experience with econometrics is desirable as is familiarity with Stata fundamentals.
|Early Registrations||Late Registrations
|Commercial / government||£1,080.00||£1,350.00|
|Academic / non-profit research||£840.00||£1,050.00|
- All costs exclude VAT, where applicable.
- Late Registrations - Registrations made within 6-weeks before the start of the course.
- Student registrations - Attendees must provide proof of full time student status at the time of booking to qualify for student registration rate (valid student ID card or authorised letter of enrolment).
- Additional discounts are available for multiple registrations.
- Cost includes course materials, lunch, refreshments and the use of computers.
The number of delegates is restricted. Register early to guarantee your place.
University of Cambridge residential accommodation is available throughout the duration of the course at additional cost. Please contact us for full information and options available. Early reservation is recommended.
Registration closes 5-calendar days prior to the start of the course.
- 100% fee returned for cancellations made over 28-calendar days prior to start of the course.
- 50% fee returned for cancellations made 14-calendar days prior to the start of the course.
- No fee returned for cancellations made less than 14-calendar days prior to the start of the course.
- Accommodation bookings are non-changeable and non-refundable.
Payment of course fees required prior to the course start date.