17  Capstone Work

Settling In

Sit with your capstone partner

  • Sign up for a meeting time with me on Thursday (see CC3 website)
  • See my Slack post about neighborhoods

Everything on the slides is in the online manual: https://bcheggeseth.github.io/452_fall_2025/

Announcements

Mini Project 3

Mini project 3 is due in 1 week

Your written communication should include the following:

  • An introduction to the historical and modern context of real estate in Ramsey County with parenthetical citations of relevant papers using bibtex
  • An introduction to your research question and why it is important
  • A discussion and justification of your model that explains WHY you made your choices
    • neighbor structure;
    • model type: OLS (if residuals are indep), SAR, CAR, Mixed Effects, Spatially Weighted;
    • mean model (choice of x’s)
  • A discussion of your conclusions based on the model and the limitations to the conclusions you can make.

Capstone Project

Expectations

https://bcheggeseth.github.io/452_fall_2025/projects/capstone-project.html

CC3

Create a new Google Document

  • Share it with your capstone group mates
  • You’ll share it with Brianna on Thursday

In the Document,

  • Draft answers for the questions in the next section

During our Conversation on Thursday,

  • Your group will discuss and “present” your answers to these questions
  • I want everyone to equally contribute

Project Proposal

Research Question/Motivation

  • What are you interested in exploring? Why?
  • What outside sources (journal articles, websites, news articles) provide context and motivation?

Data Source

  • Where will you get data to for your exploration?

Data Plan

  • What will you need to do to the data to make it usable?

Analysis Plan

  • What general steps do plan to use? What models do you plan to use?

Personal Goals

  • How do your foresee using this project to help you meet your personal goals?

Recommend Project Steps

  1. Work towards getting data “in hand”
  • Create a cleaning.R file (R Script file)
  • Read in library()
  • Read in data
  • Include all steps needed to clean existing data
  • Include all steps to create new variables
  • Include all steps to join datasets together
  • In an qmd file, you can source('cleaning.R') at the top
  1. Start by creating visualizations


  • Univariate for the outcome
  • Bivariate for outcome and explanatory variables
  • Appropriate plots for type (time/space)
    • Time Series: line plots, trend/seasonal plots
    • Longitudinal: spaghetti plots
    • Spatial: maps


  1. During EDA with visualizations


  • Keep note of strong relationships you observe (use comments)
  • Notice outliers
  • Notice missing data

  1. Start with a simple/naive model

  2. Build up to a more complex model

Today

Take 30-45 minutes to work on filling out the Project Proposal document.

Connect with your spatial partner about HW8 (due tonight) and MP3.

Wrap-Up

After Class

Before the next class, please do the following:

  • Take a look at the Schedule page to see how to prepare for the next class.