Assignments STT 3851 - Spring 2017

Jan 17:

• Go over syllabus.
• Become familiar with the Appstate RStudio server. You will use your Appstate user name and password to log in to the server. You must be registered in the class to access the server.
• Sign-up for free accounts on GitHub and Rpubs.
• When you register for a free individual GitHub account, request a student discount to obtain a few private repositories as well as unlimited public repositories. Please use something similar to first_name_last_name as your username when you register with GitHub. For example, my username on GitHub is alanarnholt. If you have a popular name such as John Smith, you may need to provide some other distinquishing characteristic in your username. Please use the same username for your account on Rpubs.
• Once you have a GitHub account, send an email to arnholtat@appstate.edu with a Subject line of STT3851 - GitHub Username, and tell me in the body of your email your first name, last name, and your GitHub username. I will then manually add you as a team member to the repository in the STAT-ATA-ASU organization that has your name (Last_name-First_name). This repository will be where you store all of your work for this course. I will also change your repository to a private repository.
• In class lecture: Review of one and two sample inference, permutation tests, and bootstrap procedures — take notes
• Homework:
  • Watch this video which provides some background on asking questions.
    
  • Read this stackoverflow thread details how to create a minimal R reproducible example.
  • Read How To Ask Questions The Smart Way by Eric Raymond and Rick Moen and heed their advice.
  • Read Getting used to R, RStudio, and R Markdown

  1. Most due dates are 5 p.m. Friday for CrowdGrader assignments.

  2. Most CrowdGrader peer reviews will be due by 11 p.m. on Tuesdays.

Jan 19:

• Before class watch the following Videos:
  • Phillip Guo’s Unix-like command line tutorials (Four tutorials about 10 minutes each)
  • Phillip Guo’s Basic Git and Github tutorials (Three tutorials about 10 minutes each)
  • Clone a GitHub Repo with RStudio (Video) (5:05)
• In class lecture: Review of one and two sample inference, permutation tests, and bootstrap procedures — take notes
• Statistics Review
• Quiz #1
• Read Chapters 1-2 of bookdown: Authoring Books and Technical Documents with R Markdown
• Link to Day One Activity and Code

Jan 24:

• Before class:
  • Read/review Getting used to R, RStudio, and R Markdown
  • Read the Git and GitHub chapter from Hadley Wickham’s book R Packages
• Follow the directions from Happy Git and GitHub for the useR to Introduce yourself to Git, Connect to GitHub, and Cache credentials for HTTPS.
• Read/review Chapters 1-2 of bookdown: Authoring Books and Technical Documents with R Markdown

Jan 26:

• Before class:
  • Watch Managing Change with RStudio (37 minutes)
• In class lecture: Review of one and two sample inference, permutation tests, and bootstrap procedures — take notes
• Statistics Review
• Schistosomiasis
• Update README file in private repository
• Use an avatar or picture on your GitHub account
• Quiz #2

Jan 31:

• Before class:
  • Read chapter 1 ISLR
  • Watch Opening Remarks and Examples (18:18)
  • Watch Supervised and Unsupervised Learning (12:12)
  • Read Section 3.4 of bookdown: Authoring Books and Technical Documents with R Markdown
• Lecture - Rmarkdown example
• Bootstrap - Rmarkdown example

• Overplotting Examples and possible Solutions

• Clone the repository to your local machine using RStudio by following these instructions:

1. Fork the repository.
2. Copy the clone URL to the clipboard.
3. Click File > New Project > Version Control > Git
4. Paste the clone URL (https://github.com/YourUserName/STAT-ATA-ASU/STT3851ClassRepo.git) in the Repository URL: box.
5. Use the name (STT3851ClassRepo) in the Project directory name: box.
6. Change if needed the location in the Create project as subdirectory of: box.
7. Click the Create Project box. You should now have a local copy of the forked repository on your local machine. Congratulations!
   

• Set the upstream remote in your fork to this repository with the command

  git remote add upstream https://github.com/STAT-ATA-ASU/STT3851ClassRepo.git

• Verify with

  git remote -v

• To obtain updates from the upstream repository type

  git pull upstream master

• If the upstream repository is using gh-pages, use gh-pages instead of master to obtain updates.

  git pull upstream gh-pages

• If there are conflicts, you will need to resolve them before proceeding.

Feb 2:

• Before class:
  • Read chapter 2 ISLR
  • Watch Statistical Learning and Regression (11:41)
  • Watch Curse of Dimensionality and Parametric Models (11:40)
  • Watch Assessing Model Accuracy and Bias-Variance Trade-off (10:04)
• Lecture - Bias Variance Tradeoff - Take notes
• Quiz #3

Feb 7:

• Before class:
  • Review/read chapter 2 ISLR
  • Watch Classification Problems and K-Nearest Neighbors (15:37)
  • Watch Lab: Introduction to R (14:12)
• Lecture - Bias Variance Tradeoff - Take notes - Slides
• Reproduce Modified Introduction to R. Submit your html file to CrowdGrader no later than 5 pm Feb 10.

Feb 9:

• Before class:
  • Review/read chapter 2 ISLR
• Go over Bias Variance Graphs
• Go over Chapter 2 Graphs
• Discuss flexible models and when to use them
• Reproduce Modified Introduction to R. Submit your html file to CrowdGrader no later than 5 pm Feb 10.
• Quiz #4

Feb 14:

• Before class:
  • Refer to bookdown: Authoring Books and Technical Documents with R Markdown as needed
• Reproduce Writing Assignment. Submit your html file to CrowdGrader no later than 5 pm Feb 17.

Feb 16:

• Work Day
• Reproduce Writing Assignment. Submit your html file to CrowdGrader no later than 5 pm Feb 17.
• Quiz #5

Feb 21:

• Before class:
  • Read pages 59-82 of ISLR
  • Watch Simple Linear Regression and Confidence Intervals (13:01)
  • Watch Hypothesis Testing (8:24)
  • Watch Multiple Linear Regression and Interpreting Regression Coefficients (15:38)
• Discuss Chapter 2 and 3 material

Feb 23:

• Before class:
  • Read pages 82-120 ISLR
  • Watch Model Selection and Qualitative Predictors (14:51)
  • Watch Interactions and Nonlinearity (14:16)
  • Watch Lab: Linear Regression (22:10)
• Discuss Chapter 2 and 3 material
• Go over Bias Variance Graphs again
• Quiz #6

Feb 28:

• Answer questions at the end of Chapter 2 and 3 material.
• Reproduce Lab: Linear Regression. Submit your html file to CrowdGrader no later than 5 pm Mar 3.

Mar 2:

• Before class:
  • Start Project 1 - Read the hyperlinked paper.
    
  • The data set can be downloaded from http://www.amstat.org/publications/jse/datasets/homes76.dat.txt, while the code book is available from http://www.amstat.org/publications/jse/datasets/homes76.txt
• Reproduce Lab: Linear Regression. Submit your html file to CrowdGrader no later than 5 pm Mar 3.
• Quiz #7 (Chapter 3 Assessment)

Mar 7:

• Work on Project 1
• Code for class notes

Mar 9:

• Work on Project 1
• Quiz #8 (Chapter 3 Assessment)

Mar 13 - Mar 17: No Class - Spring Break

Mar 21:

• Commit your final changes for Project 1 to your private repository and email the instructor your SHA prior to 10am March 22, 2017. Use the subject line “Project 1-STT3851-Last_name First_name-SHA#” in your email.

Mar 23:

• Before class:
  • Read chapter 5 ISLR
  • Watch Estimating Prediction Error and Validation Set Approach (14:01)
  • Watch K-fold Cross-Validation (13:33)
  • Watch Cross-Validation: The Right and Wrong Ways (10:07)
• Discuss Cross-Validation Hand Out
• Quiz #9

Mar 28:

• Before class:
  • Read/re-read chapter 5 ISLR
  • Watch Lab: Cross-Validation (11:21)
• Discuss Cross-Validation Hand Out
• Reproduce Lab: Cross-Validation

Mar 30:

• Before class:
  • Watch Linear Model Selection and Best Subset Selection (13:44)
  • Forward Stepwise Selection (12:26)
  • Backward Stepwise Selection (5:26)
• Finish Cross-Validation Hand Out
• Complete Reproduce Lab: Cross-Validation. Submit your html file to CrowdGrader no later than 5 pm March 31.

Start Project 2

• Example Template Directory
• Use the housedata.csv file - See King County Housing Data
• Quiz 10

Apr 4:

• Before class:
  • Watch Estimating Test Error Using Mallow’s Cp, AIC, BIC, Adjusted R-squared (14:06)
  • Watch Estimating Test Error Using Cross-Validation (8:43)
• Notes on Credit Problem

Apr 6:

• Before class:
  • Watch Shrinkage Methods and Ridge Regression (12:37)
  • Watch The Lasso (15:21)
  • Watch Tuning Parameter Selection for Ridge Regression and Lasso (5:27)
• Quiz #11

Apr 11:

• Before class review Book Down
• Before class
  • Watch Dimension Reduction (4:45)
  • Watch Principal Components Regression and Partial Least Squares (15:48)
  • Watch Lab: Best Subset Selection (10:36)
  • Watch Lab: Forward Stepwise Selection and Model Selection Using Validation Set (10:32)
  • Watch Lab: Model Selection Using Cross-Validation (5:32)
  • Watch Lab: Ridge Regression and Lasso (16:34)

Apr 13:

• Work on Project 2
• Work on Lab: Subset Selection Methods
• Quiz #12

Apr 17-18: No Class - State Holidays

Apr 20:

• Work on Project 2
• Work on Lab: Ridge Regression and the Lasso

Apr 25:

• Turn in Project 2 - NLT April 26, 10 am
• Make sure a single .Rmd with the sections outlined in the Example Paper is in your Project2 directory under your private course repository.
• Send an email to the instructor using a subject of “House Data Predictions” with an attached vector (csv file) of your predictions using the data set housedataTEST.csv. Note: your vector should contain 4229 predictions.

Apr 27:

• Poster Template Directory
• Work on Presentation
• Quiz #13

May 2:

• Work on Presentation

May 11:

• Final - Presentations 9:00-11:30

Home Page - Syllabus - Piazza