---

Applied Regression Analysis (BUS 41100)

This website is ONLY for Autumn 2020, the remote version of 41100. For course material from previous quarters, go here.

Instructor: Max H. Farrell

Office hours (over Zoom): Wednesday 4:30 – 6PM, Saturday 11:30AM – 1PM

TA: Gustavo Gonzalez

Syllabus

Course material:

1. This website: for all slides, homework, and data sets
2. Piazza for Q+A: your first stop for help
3. Zoom: all lectures will be over zoom, links will be sent in advance to students, and recordings will be available
4. No Canvas site

Before class starts:

1. Homework Zero – To test your readiness for this course
2. Get started on R before class: see Computing below

Notices

• Syllabus updated: version 1.2 has new course schedule, including the midterm!
• Syllabus updated again: version 1.3.1
• New version of homeworks 5 and 6 posted

Lectures

These will be updated as we go along, so always download the latest version

• Week 1: Introduction, Simple Linear Regression (SLR)
  • Slides | R Code | Data: pickups and sales
• Week 2: Inference for SLR
  • Slides | R Code | Monte Carlo Code | Data: mutual funds and SAP ROE
  • Handout: deriving variance formulas and explaining prediction intervals
• Week 3: Finish SLR, Start Multiple Linear Regression (MLR)
  • Slides | R Code | Data: mutual funds, telemarketing, and sales
  • Handout: polynomial regression & keeping lower order terms | R Code | Data: telemarketing and wages
• Week 4: Multiple Linear Regression Part II
  • Slides | R Code | Data: price experiment, bike rentals, pickups, grades, GDP & imports, and canned goods
  • Handout: interpreting log() models
• Week 5: MLR Pitfalls and Some Fixes, Clustered and Panel Data
  • Slides | R Code | Data: telemarketing, teacher salaries, pickups, Patents and R+D
• Week 6: An Introduction to Time Series
  • Slides | R Code | Monte Carlo Code | Data: airline, beer, Dow Jones Industrial Average, and weather
• Week 7: Logistic Regression
  • Slides | R Code | Data: pricing experiment, NBA point spreads, and German credit
• Week 8 and Week 9: Classification and Model Building 1
  • Slides | R Code | Data: census, NBA point spreads, and German credit

Homework

• Homework 0  |  No data required, no solutions available
• Homework 1  |  box plots, scatter plots, stock market
• Homework 2  |  Monte Carlo code | SMSA data
• Homework 3  |  teacher salaries, tractors
• Homework 4  |  pricing experiment, cheese, Census data
• Homework 5  |  newspapers, NSW+PSID, beef
• Homework 6  |  UK gas consumption, US gas price, crime, bike sharing, pricing experiment, Grunfeld

Handouts

• Deriving variance formulas, including leverage, and explaining prediction intervals
• Polynomial regression & keeping lower order terms | R Code | Data: telemarketing and wages
• Interpreting log() models

Computing

The default computing language for this course is R, is free (as in speech) and available from CRAN. Other languages (e.g. python, MATLAB, STATA, ...) are allowed. Examples in lecture, homework solutions, etc., will be in R.

Get started before class starts!

• A good introduction/tutorial to R is here.
• UCLA has a fantastic help page for R (and statistics/regression in general) with everything from installation/basic help, worked-through examples, books, and link to more resources.
• The University offers R workshops in the Research Computing Center, see schedule here and has e-books available here.
• The resources out there are continually changing, so you may find other options. Please let me know if you find something helpful that isn't listed here.