About econpy.org

econpy.org is my take on some of the ways Python can be leveraged in the typical research process that an economist or basically any other empirical researcher follows. Most of the material on econpy.org focuses on the use of Python to obtain structured and unstructured data from the Internet. The Internet is packed with tons of high quality data just waiting to be combined in the right way. However, academics (especially in non-computational fields, like the social sciences) often find themselves discovering unstructured data on the Internet that they want to analyze for their research, but aren't sure how to collect it. Enter Python.

Having great data, and a lot of it, is a wonderful way to improve an empirical research study. But what is great data? It's similar to the big data buzz word that's currently floating around, however big data refers to not just having more data, but also the ability to measure things at a much finer level. For economists, another buzz word for this idea is nanoeconomics.

Overall, having a larger sample filled with observations that control for much more detailed features not only reduces the chance that you'll have to use more "risky" statistical estimators, but it's also a step towards making the data your drawing inferences from a better representation of the actual world. Not only that, but if you create your data set yourself, then you'll be the only one with that same data set. One implication that follows from the ability to engineer your own data is that research design will become an extremely important characteristic of your study.