"If economists could manage to get themselves thought of as humble, competent people on a level with dentists, that would be splendid." – John Maynard Keynes (1930)
It's hard to be humble when your models provide interesting insights into human behavior. By helping us understand how the incentives that we face influence the decisions that we make, economics provides a good toolkit for measuring those incentives and their effects on our decisions.
But economists must also remember that economic theory has its limitations. For example, astronomers recently photographed a black hole. In economics, there is no black hole to take a photograph of.
Our mathematical models can yield insight into the incentives that we face and the decisions that we make, but those models assume that we are always optimizing. In practice, humans usually make decisions by following rules of thumb, so behavioral economics has been very successful because it incorporates psychology into its models.
Economics is also a good place to begin exploring other fields too. For example, my knowledge of economics helped me develop a Sicilian dictionary and machine translator and because it taught me the mathematical and statistical tools used in natural language processing. The project has been a lot of fun, and so I hope that economics will be useful to you in other fields too.
When I created this website in 2002, it was just a simple way to organize my research and teaching materials. But as I kept adding new courses, this site grew, so I created wdowiak.me for my research and photography. Later, I created Napizia for my Sicilian language projects.
Today, this site still houses my economics courses. My micro lecture notes and macro lecture notes have been particularly popular. But I wrote them at a time when students still used paper. Today, phones have put learning into students' hands, so I am rewriting my notes in HTML. Check out my MathML and SVG demo.
And as computer technology has evolved, the field of economics has too. The regression models used in econometrics are similar to the neural networks used in machine learning. Both estimate model parameters by minimizing a loss function. And the estimated models are used to make predictions. In econometrics, a model might predict the unemployment rate. In machine learning, a model might predict what object appears in a picture. The predictions are very different, but statistical foundations are the same.
To help you estimate your own models, this website also provides notes on computer programming. My notes on the Perl language show you how to assemble a dataset. And my notes on the R language show you how to conduct a statistical analysis.
The common theme on each of these pages is optimization. As a matter of mathematics, there is no difference between profit maximization, linear regression and machine translation. They are all examples of optimization.
In microeconomics, the firm seeks the quantity of labor that maximizes its profit. In regression, we seek the parameter estimates that minimize the sum of squared residuals. And in machine learning, we train our model by minimizing a loss function. Each is an example of mathematical optimization.
Writing all of the content for this site was no small effort. I wrote it for my students and I wrote it for you. Please respect the work that I created for you. You are welcome to include in your own work short quotes from this website; all I ask in return is proper attribution. Below is a sample citation:
"Just Split, Dropout and Pay Attention." economics @ doviak.net.
This website records your IP address and your browser's user agent. That's the only personal infomation that this website collects. This website does not store any cookies in your browser.
But we would like to keep you informed of the new content that we are developing here, so we hope you will subscribe to our mailing list. If you do, we will keep your personal information secure. We do not sell our mailing list to anyone. And we will not flood your inbox with mail. It takes a lot of time to develop this material, so we do not send mail frequently.
Thank you for visiting doviak.net. I hope you enjoy my courses as much as I enjoyed creating them for you. If you have any questions or if you would like to contact me, you may email me at: email@example.com.
Copyright © 2002-2023 Eryk Wdowiak