For details about the final project like what you should hand in, and how you should structure your reports, see Final project details.

The basic idea behind the final project is simple: pose an interesting question; collect a relevant data set; and use the data, in conjuction with the statistical modeling tools we have learned in class, to answer the question you have posed. Make sure to quantify any uncertainty that arises in answering your question, and to address any shortcomings in the answer provided by your data and analysis.

You can see some example project descriptions here: Final project examples. See also the gas prices case study in week 12, which is based on a previous student project

This assignment is purposely open-ended, allowing you considerable freedom to follow a path dictacted by your own intellectual curiosity. You will be evaluated both on the technical correctness and the overall intellectual quality of your presentation. A thoughtful analysis of a well-posed question using a carefully chosen dataset and appropriate methods is more important than engaging in fancy analysis or complicated data collection. That said, creative data collection efforts will be rewarded!

Where can one find data? Everywhere! These websites in particular are good resources:

• https://datasetsearch.research.google.com/
• https://github.com/caesar0301/awesome-public-datasets
• http://stats-for-change.github.io/data.html
• https://www.propublica.org/datastore/datasets

If you want to branch out even further, here’s a short list of other sources you might consider: major newspapers, the U.S. Census, the Federal Reserve, academic journals, the Economist, Twitter, the World Bank, ESPN.com or other sports sites, Craigslist, Amazon prices, EBay, the Bureau of Labor Statistics, Facebook, the World Economic Forum, the OECD Factbook, the CIA World Factbook, the Securities and Exchange Commission, Yahoo Finance, Google Public Data Explorer, your own vital signs, your own experiment or survey, your favorite blogs, your other classes, and your friends. If you know how to write code that will scrape a website, your options are almost limitless here.