What is Econometrics?

and what will we be learning in ECON 4750?

Brantly Callaway

What will we learn this semester?

\(\newcommand{\E}{\mathbb{E}}\)

At a very high level, our goal for this semester is to develop skills related to using data to learn something

Two main course objectives:

  • Prediction

  • Causal inference

Prediction

Examples:

  • credit risk / default probabilities
  • product recommendation systems
  • demand prediction

Tools:

  • modern approaches to model selection
  • machine learning

Causal Inference

Motivation: correlation is not causation

  • pirates and global warming
  • police and crime
  • going to the doctor and illness

Running experiments is a a leading solution to this problem

  • chemists can exogenously manipulate temperature on a Bunsen burner
  • in medicine, it is common to randomly assign some patients to a new drug and others to a placebo
  • some examples in economics: job trainging, conditional cash transfers, many in developing countries

Causal Inference

But experiments are often not feasible

  • interest rates
  • tax policy
  • people’s education levels
  • minimum wages

We will learn a number of tools/tricks for causal inference with observational data

We will also learn to be careful with our own causal claims and think carefully about others’ causal claims

  • There may be (a large number of) cases where we have relevant data but still can’t fully understand causal relationships between our variables of interest

Tenor of the Course

We will have the dual goals of learning how to do prediction/causal inference/etc. and how it works

Along the way…

We will go in depth on a number of a related topics/tools

  • Statistical programming

  • Probability and statistics

  • Linear regression

These are the “language” we will be speaking this semester, and I aim for you to be experts (or at least very familiar) on all of these topics by the end of the semester.