Mini Project 2

GLMs

Evening Before (Wednesday 9/4)

Pick a partner.
Find an appropriate data set.
- GLM’s are broadly applicable and there are no requirements on the type of data the response needs to follow.
- The data you choose should have a few predictors that may be useful when modeling (ideally at least 1 quantitative and 1 categorical).

Data Options. Check out Data Links on the Useful Links part of the course website. TidyTuesday has quickly accessible data.

Write out the anticipated cleaning and/or feature engineering steps you will need to take. Some examples:
- Creating simplified categorical variables or transforming a continuous variable into categorical.
- Aggregating data
- Converting date columns
- etc.
Setup your R Project(s) on the server.
Read your data into R. This likely will require you to download the data to your computer and upload the data the server.

This project will be in a workshop style. The intention is for you to start and finish by the end of class time. We will follow a timeline:

Task	Timing
Clean Data	9:00 am - 9:30 am
Perform EDA	9:30 am - 10:30 am
Fit, Assess, and Compare Regression Models	10:30 am - 11:00 am, 1:00 pm - 2:00 pm
Prepare presentation	2:00 pm - 2:20 pm
Present your findings	2:20 pm - 2:40 pm
Submit your Final Report	Submit HTML Sunday 9/8 at 11:59 pm

Each Mini Project is worth 50 points (Labs are 10 points each).

Category	Points
The data chosen is appropriate, and the cleaning steps are correct and explained.	5
EDA is thorough. All included graphs and tables are paired with a discussion. EDA supports the choice of modeling technique.	15
The model fitting process has a logical flow. Multiple models are considered and compared using statistical tests and multiple metrics. Any model that is interpreted has been assessed using residual plots and appropriate statistical tests.	15
The code follows a sensible order and has been appropriately commented on.	5
The presentation is concise, describes the data, highlights key parts of the EDA, describes minimally the final model, gives at least 1 interpretation in the context of a coefficient, and discusses limitations and potential future work.	5
The report is well written, with correct spelling and grammar. The used code is included either inline or in an appendix at the end.	5

Add format part of your final report document and then re-render:

```{r}
#| label: yaml_example
#| eval: false
---
title: "Document title"
author: "my name"
format:
  html:
    embed-resources: true
---
```

When you are finished with your homework, be sure to Render the final document. Once rendered, you can download your file by: