1
$\begingroup$

I am an undergraduate student and am planning to take a course in Regression Analysis next semester whose description is

"The analysis of continuous response data. The focus is on linear and multiple regression with theoretical and practical training in statistical modeling. A hands‐on, applied course where students will become proficient using computer software to analyze data drawn from a variety of fields, and will learn what assumptions underlie the models, how to test whether the data meet the assumptions, and what can be done when the assumptions are not met. "

I am signing up for it because the university I attend currently does not offer a course in Econometrics and Regression is the closest thing to Econometrics that is offered. I want to understand Econometrics at the level of Stock and Watson's Introduction to Econometrics and/or Jeffrey Woolridge's Introductory Econometrics textbook.

Can someone please explain what the differences might be between the Regression course mentioned above and a course that would be taught out of the Econometrics textbooks mentioned above? If in a hypothetical world, I could take a course next semester supplementing the Regression course to patch up my understanding of Econometrics, what book/ source would you recommend that I read? Any advice is greatly appreciated!

Aside: I want to get acquainted with the world of economic research and eventually pursue a doctoral degree in Economics.

Improve this question
$\endgroup$

2 Answers 2

Reset to default
7
$\begingroup$

To put it simply regression modeling is a technique, a tool, used by econometricians. Regression deals with dependence amongst variables within a model. But it cannot always imply causation.

As economists, we are biased towards establishing causal relationships. So, this explains our focus on endogeneity issues and strategies of identification to establish causality.

In other words, as Ragnar Frisch (one of the founders of the Econometric Society) would say

Econometrics by no means the same as economic statistics

See also the answers to What is an econometrician? where the key idea is that economic theory is crucial to econometrics and understand measurement. Neither theory nor measurement on its own is sufficient to further our understanding of economic phenomena. We need both and measurement without theory is unlikely to provide a satisfactory explanation of the way economic forces interact with each other.

In summary, regression is super useful but not sufficient to make causal inference and to understand economic phenomena.

Stock and Watson's Introduction to Econometrics and Jeffrey Woolridge's Introductory Econometrics are both excellent textbooks.

Improve this answer
$\endgroup$
3
  • 1
    $\begingroup$ That was a very nice differentiation between econometrics and general regression. One other important difference is that econometrics often deals with modelling behavior. Regression can be used as a tool for doing this but there are additional issues involved ( that I won't get into here ) that statisticians usually don't get involved with. Generally speaking, statisticians are not dealing with behavioral issues. $\endgroup$
    – mark leeds
    Apr 18, 2020 at 0:11
  • 1
    $\begingroup$ This is a fantastic answer. Also, at least in my econometric training, there was a strong mathematical theory element that doesn’t appear to be present in this course description. $\endgroup$
    – dismalscience
    Apr 18, 2020 at 0:22
  • 2
    $\begingroup$ As a statistician who has to consider results of econometric studies, the causal conclusions are more often than not less than convincing. What is often more impressive and special is the handling of time series and autoregression to avoid spurious correlation, though when done well this tends to lead to weaker results $\endgroup$
    – Henry
    Apr 18, 2020 at 1:16
2
$\begingroup$

Econometrics at my college covered some time series analysis related stuff as well. The main difference was that the business course Regression analysis taught regression as a tool to establish causal relationship while Econometrics dived into the mathematical reasoning, potential flaws i.e. multicollinearity, spurious regressions, and techniques on how to correct them. I wanted to take regression analysis after I had taken Econometrics but my advisor told me it would be like taking basic Algebra after Calculus. I would recommend Woolridge's book to do on your own.

There are also many online resources available. Look into MIT opencoursware. They have an entire lecture series for Econometrics with assignements, quizzes, exams, and solutions available... Free of Cost.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.