What This Document Is
This document presents a detailed exercise focused on applying statistical hypothesis testing within the framework of a multinomial logit (MNL) model. It’s designed for students studying advanced econometrics, specifically those delving into discrete choice models. The exercise utilizes an artificially generated dataset concerning travel mode selection – considering options like bus, car, and staying home – to illustrate practical applications of maximum likelihood estimation. It builds upon lecture material and aims to solidify understanding through hands-on application.
Why This Document Matters
This resource is ideal for students in an introductory econometrics course (at the graduate level) who are looking to strengthen their ability to translate theoretical concepts into practical analysis. It’s particularly beneficial when preparing for assignments or exams that require applying statistical tests to MNL models. Students who want a deeper understanding of Wald, Likelihood Ratio (LR), and Lagrange Multiplier (LM) tests in the context of discrete choice modeling will find this exercise valuable. Access to the full document unlocks a complete, step-by-step guide to performing these tests.
Topics Covered
* Multinomial Logit Modeling
* Maximum Likelihood Estimation (MLE)
* Hypothesis Testing (Wald, LR, LM tests)
* Discrete Choice Analysis
* Statistical Inference
* Data Analysis using SST or TSP software
* Travel Mode Choice Modeling
What This Document Provides
* A detailed description of a discrete choice scenario with a generated dataset.
* A framework for conducting statistical tests on MNL model parameters.
* Background information on the application of various hypothesis testing methodologies.
* A reference to specific software (SST) and the possibility of transcription to alternative software (TSP).
* Output from an example run to guide the user through the process.
