What is it about?
Maximum likelihood is the statistical method of choice for estimating parameters in situations where data are not amenable to the use of statistical techniques that are generally applied to data from experimental studies with a homogeneous population (such as difference-in-differences analysis). A well-specified economic model is a useful tool for understanding the consequences of policy changes that are not necessarily marginal or that affect the distribution of outcomes as well as the mean outcomes. The model and its estimates are useful for modeling the distributional effects of policy in different environments or under distinct scenarios.
Featured Image
Why is it important?
Data is not generally generated by a well constructed experiment. This does not mean that the data should not be used to obtain useful measurements. Maximum likelihood and an economic model can be used together to obtain useful estimates.
Perspectives
This was harder to write than I initially thought it would be...whether it is easy to read...I don't know. I hope in the end that it provides an introduction and some reasons to look in the direction of more 'structural' modelling.
Professor Gauthier Lanot
Department of Economics, Umeå School of Business University, Umeå University
Read the Original
This page is a summary of: Maximum likelihood and economic modeling, IZA World of Labor, January 2017, Institute for the Study of Labor (IZA),
DOI: 10.15185/izawol.326.
You can read the full text:
Contributors
The following have contributed to this page
