Skip to Main Content
It looks like you're using Internet Explorer 11 or older. This website works best with modern browsers such as the latest versions of Chrome, Firefox, Safari, and Edge. If you continue with this browser, you may see unexpected results.

ECON 4370: Econometrics

Data sData sources and sources for scholarly work for ECON 4370: Econometrics course

Welcome and General Information

Welcome to the library guide for ECON 4370: Econometrics course. This guide is meant as a starting point and a resource for you as you work on assignments related to econometrics.

Please feel free to contact Alex Gallin-Parisi, liaison librarian for economics, with any questions about your research or compiling your final paper.

"What distinguishes an econometrician from a statistician is the former's preoccupation with the problems caused by violations of statisticians' standard assumptions; owing to the nature of economic relationships and the lack of controlled experimentation, these assumptions are seldom met." --Peter Kennedy, A Guide to Econometrics, Fifth Edition (2003), p.1

A word to the wise (tips and thoughts)

Suggestions for Trinity students in this course from the Senior Associate Director of Division of Research and Statistics, Board of Governors of the Federal Reserve System:

  1. Always graph your data.  It’s good for you and for the reader.  Look for outliers, weird patterns, etc.  You don’t want to blindly follow textbook stats models.
  2. But don’t just trust your eyes either!  You can be easily fooled into seeing patterns that aren’t real.  That’s what the stats are for.
  3. Lots of data get revised.  For example, the employment and GDP numbers you look at now for, say, 2007 are not the numbers policymakers were looking at in 2007.  This can matter a lot for analysis of policy decisions that were taken “in real time.”  So, for some work you should look at “real-time data.”  See Real-Time Data Set for Macroeconomists (philadelphiafed.org)
  4. Be careful with “controls.”  In work on wage inequality, a common thing to do with wage equations used to be to regress/log wages in education, work experience, industry, a whole bunch of other things, and, say, a dummy variable for gender. But if discrimination drives gender difference in “choices” for labor force participation (and therefore work experience), education, industry, or occupation, you will understate the effect of gender difference on wages.  “Controlling” for, say, industry, actually throws away some of the interesting and important variation you are interested in.  So, you need to look more generally at the patterns.  This applies in many many areas of study.  The general thing to ask yourself is, what is your research question and how does it affect how you should think about controls.
  5. Take the time to read the data descriptions. It can be boring. But you should know what you are working with.  Sometimes there are weird subtleties that can affect your approach or your interpretation.  Be particularly careful of “data source telephone.”  Someone writes about data in a paper, someone else reads that description and writes another paper, and describes the data a bit differently, and so on.  A few links down the chain, the description can be wrong, or at least off. [This really happens.]