The Robustness of Limited Dependent Variable Modles to Misspecification

The purpose of this research project is to apply Monte Carlo simulation to test the robustness of results from Limited Dependent Variable (LDV) models to misspecification, particularly as applied by empirical economics papers. In particular, we will investigate the robustness of parameters and marginal effects to a violation of the model’s assumptions.

For linear and linear mixed models, the literature suggests that p-values “are highly robust to even extreme violation of the normality assumption” (Knief and Forstmeier, 2020). This generalisation is not true of LDV models, where the maximum likelihood estimates can be inconsistent under non-normality (Bera, Jarque, and Lee, 1984).

Luke Thomas

The University of Western Australia

Luke Thomas is a third-year Bachelor of Philosophy student at the University of Western Australia, majoring in mathematics and economics. Luke has worked as a research assistant at the Centre for Social Impact, and is interested in how computational techniques and data can be applied to evidence-based policy and decision-making.

Luke is excited to explore the robustness of statistical models used in economics over the 2020/21 AMSI VRS program.

You may be interested in

Alex Oakley

Alex Oakley

Appraisal of Machine Learning Evaluation Techniques in the Context of Glacier-Generated Events
Montana Wickens

Montana Wickens

Optimal Control of Translocation Strategies for Threatened Australian Mammals
Susanna Grigson

Susanna Grigson

Hierarchical Classification of Protein Sequences
Sean McGowan

Sean McGowan

PDE Discovery Using Machine Learning
Contact Us

We're not around right now. But you can send us an email and we'll get back to you, asap.

Not readable? Change text.