I firmly believe that the best way to learn about statistical modeling is to do statistical modeling. Every time I run a course like this one, I use a Kaggle Competition and corresponding series of Competition Assignments to motivate students to go further and learn more about statistical modeling techniques. This is a friendly competition associated with our class (and perhaps similar classes at institutions like our own). To be clear, your ultimate course grade is not tied to your position on the leaderboard. However, students in these courses often mention that the Competition Assignments are motivating and one of the most valuable parts of the class experience. I hope you’ll all find this as well!

Complete the following:

  1. Check the #competition-questions channel in Slack for a direct link to our Kaggle Competition for this semester.

  2. Peruse the competition site, reading the details for this semester’s competition. On the Overview page for the Competition, click on the black Join Competition button.

  3. Navigate to the Data tab on the competition site. Download the data.csv and comp.csv files. You can also download the sample_submission.csv file if you like, but you don’t need to. Put these files into a convenient location on your computer.

    • You may be prompted to accept the competition rules before you are allowed to download the data. The rules are to (i) learn new things, (ii) don’t cheat, (iii) apply yourself, and (iv) have fun.

  4. Open RStudio and use File -> Recent Projects to select and open the R Project which is managing your GitHub repository.

  5. Use File -> New File -> R Markdown... to create a new R Markdown notebook. Fill in the fields as usual and then click the button to create the notebook.

  6. In the setup chunk, add the code necessary to load the {reticulate} library. You may also want to load {kableExtra} for pretty table printing.

  7. Open a new Python code chunk and import pandas.

  8. Also in that Python code cell, use pd.read_csv() with the path to the data.csv file on your computer to read that data into a named object.

    • If you are having trouble identifying the file path, you can click on the Environment tab in the top-right pane of RStudio and click the Data Import button. From here, you can click on Browse and navigate to the location where you saved the file. Don’t completely read in the data this way though. Look in the lower right to see the code necessary to read in your data. Part of the second line in that code viewer will be the path to you file. Copy it and paste it into the read_csv() function in your notebook. As we’ve done with URLs, you’ll need to surround the file path with quotes.

  9. Remove the boilerplate that appears after your setup code chunk.

  10. Add a new code cell and use it to print out the .head() of the data you’ve just read in.

  11. Use the blue ball of yarn button to knit the notebook into an HTML document.

  12. Use the Git tab in the top right pane of RStudio to Pull, Commit, Push your new files to your remote repository at GitHub.

Stop by my office if you have any questions or need help.