Module Specifications
Academic Year 2026 - 2027
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Description This module will develop students’ understanding of econometric theory and enhance their ability to work in practice with data types commonly encountered in Economics. Students will review the basics of linear regression, including practical issues such as the choice of functional form and interpretation of results. They will learn how to test and correct for non-adherence to Gauss-Markov assumptions in multiple regression. The module will extend students' skills to include analysis of panel data, discrete choice, and survival analysis. Common causal methods used in microanalysis, such as Differences in Differences, Regression Discontinuity Design and Matching models will be addressed, and students will be introduced to machine learning techniques using Lasso and Ridge regression. The module’s emphasis on practical applications will give students experience in all aspects of an applied research project, from formulating research hypotheses and choosing appropriate data and econometric analysis to interpreting and critically evaluating results. Students will be expected to engage in lectures, seminars and practicals, familiarise themselves with statistical software, complete take-home and in-class assignments, and contribute to an independent group-based practical project. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Learning Outcomes 1. Explain the underpinnings of Ordinary Least Squares and Maximum Likelihood estimation, including the conditions under which MLE and OLS produce equivalent regression estimates. 2. Discuss the Gauss Markov Theorem and the theoretical assumptions required for BLUE estimators in OLS 3. Identify and correct for common problems in practical regression analysis and issues arising from non-adherence to Gauss Markov assumptions. 4. Select the appropriate econometric techniques to analyse discrete dependent variables and time-to-event data and interpret and evaluate results. 5. Use panel data to determine causal relations. 6. Articulate the fundamentals of regularisation algorithms applied to regression analysis and the application of Ridge and Lasso techniques. 7. Apply modelling techniques addressed in class using STATA or another econometrics programme. 8. Collaborate to share ideas; discuss, negotiate and reach concensus; and manage workload to produce a reasoned, coherent report. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
All module information is indicative and subject to change. For further information,students are advised to refer to the University's Marks and Standards and Programme Specific Regulations at: http://www.dcu.ie/registry/examinations/index.shtml |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Indicative Content and Learning Activities
Lectures: Underpinnings of Multivariate Regression and Inference. Review of the fundamental theory underlying Ordinary Least Squares regression analysis assuming a continuous dependent variable. Deriving OLS estimates and standard errors. Hypothesis tests. Goodness of fit. Lectures: Common issues in multiple regression: Discussion of the Gauss Markov assumptions necessary to provide BLUE estimates; the impact of non-adherence to standard OLS assumptions on parameter estimates and standard errors, and implications for inference, how to detect whether assumptions hold, and procedures when they do not. This section will address the following issues: (i) Omitted variables (ii) Multicollinearity (iii) Heteroscesdasticity (iv) Correlated errors (v) Endogeneity concerns and use of Instrumental Variables Lectures: Method of Moments, Generalised Method of moments, Maximum Likelihood Estimation (MLE) Placing OLS, GLS, IV in the context of MM and GMM. Explanation of MLE in the context of regression analysis, equivalence of MLE and OLS in the context of regression with normally distributed errors. Lectures: Discrete Choice Analysis Modelling (i) bivariate outcomes using logit and probit. (ii) ordinal responses using ordinal logit, including consideration of generalised ordered logit (iii) multiple choice data using multinomial logit, including inference and critiquing results, with applications to business contexts. Lectures: Survival analysis Time-to-event data modelling, with application to the job and housing markets. Lectures: Panel Data Models for panel data: Fixed and random effect models; Use of difference in differences to determine causal effects (simple 2x2 design revisited, multiple groups and time periods, TWFE, hetrogeneity, pre-treatment trends, staggered adoption). Lectures: Regularised Models Use of regularised models: Lasso and Ridge regression Lectures: Additional Causal Techniques Matching methods (Exact matching, Propensity Score Estimators, Coarsened matching); Regression Discontinuity Design (Non-parametric regression, Sharp Regression Discontinuity, RDD with covariates, Fuzzy Regression Discontinuity) Seminars Seminars will run during the second half of semester. During seminars, groups will present their project ideas and preliminary project findings and receive feedback and advice from the module lecturer and fellow students. Computer Practicals Using STATA Introduction to STATA: inputting data, summary statistics, plots and correlations, basic hypothesis tests (e.g. difference in means). Regression analysis: OLS Regression, G-M checks, IV Panel data models, Discrete choice models, Lasso and Ridge regression. Students will work in a controlled, supervised environment to ensure required engagement with software. This will include supervised in-class work on group practical projects. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Indicative Reading List Books:
Articles: None | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Other Resources
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||