BA 355: Business Analytics {Fall 2021}

General Course Material:   

• Syllabus
• Canvas

---

Notes:

· Tu 10/26: Discuss outliers, ICE 5,Case 3.1

· Th 10/28: Discuss Multiple LR, work on cases.

· Tu 11/2: Work on Cases 3.1 & 3.2.

· Th 11/4: Asynchronous: Work on Case 3.2, get at least half way done… I will be available in my office (EBH 158) and on Zoom – send me an email and I’ll open up a Zoom room.

· Tu 11/9: Start Case 4: ZIP Codes

· Th 11/11: Case 3.2 and Case 4

· Tu 11/16: Work on Case 4 in class, decision on what’s due before TG Break.

· Th 11/18: Case 4

· Tu 11/30: Introduce Case 5, work on Cases 4 and 5.

o Rest of the term: Case 4 and Case 5 (and possibly one more assignment)

· Th 12/2: Complete Case 4, start Case 5

· Tu 12/7: Meet in class. Course wrap up. Work on Case 5.

· Th 12/9: Complete Case 5, due on Tu 12/14.

Intro to BA:

• ICE 1
• Bubbles Data -- with corrected data.
• ICE 2
• Assignment 1
• http://www.ispot.tv/ad/7Fip/ibm-watson-analytics-make-smarter-decisions-feat-dominic-cooper

Case 1: NFL Point Spreads

· Case 1 Part 1

o NFL Data for 2013 - 2016

o Point Spreads and Win Percentages

o Logistic Curve fitted with Solver

· As published in INFORMS Transactions on Education (link)

· What to do with four ties from 2013 – 2016? For underdog, a tie is a win? https://en.wikipedia.org/wiki/1968_Yale_vs._Harvard_football_game

• Stern Paper
• On the (Non-Linear) Probability of Winning
• What is home field advantage worth?
• These are videos I made last year when we were on semi-lockdown. We will cover the same material in class on Thursday 9/2 and Tuesday 9/7 so you probably don’t need to watch them, but here they are just in case. Case 1 Video 1, Case 1 Video 2, Slides
• Curve Fitting: ICE 3, ICE 3 Data
• Case 1 Part 2

*** Answer Key ***

Case 2: Credit Scores and Loan Default Rates

· Assignment 2

· Case 2.1, Credit Score Data

· ICE 4

o Answer Key, Excel

· Case 2.2, Data

o Answer Key, Excel

Case 3: Zillow/Real Estate/Outliers

· Case 3.1

· ICE 5

o Outliers file from class on 10/26

· Multiple Linear Regression Example

· Case 3.2

o Clean Data

o Answer Keys: Word, Excel

· Assignment 3

o Raw Data

Case 4: Zip Codes

• Original Request
• FLC Zip Code Data
• Case 4
• Distance Formula: http://bluemm.blogspot.com/2007/01/excel-formula-to-calculate-distance.html
• Database source 1: http://federalgovernmentzipcodes.us/
• Hints:
• =vlookup() – make sure data are sorted by zip code
• Straight line distance estimates can be found from lat and long
• Map function in Excel
• Case (As published in INFORMS Transactions on Education)

Case 5: Ska Brewing

• Case 5 (Updated!)
• Data
• Original Case (As published in The Applied Business Analytics Casebook: Applications in Supply Chain Management, Operations Management, and Operations Research)

---

Older Notes:

· Tu 8/31: Slides. Intro to Analytics, ICE 1

· Th 9/2: Intro to Case 1, ICE 2

· Tu 9/7: Start Case 1

· Th 9/9: More Case 1

o What is best linear model?

o Look at links…

o What should predictive model look like?

· Tu 9/14: Fit Logistic Curve to data.

o Logistic Curve:

· Th 9/16: ICE 3, ICE 3 Data

· Tu 9/21: Start Case 1.2 – Data Cleaning

· Th 9/23: Start Case 1.2 – Data Cleaning

· Tu 9/28: Start Case 2.1

· Th 9/30: Finish Case 1.2, Start Case 2.1

· Tu 10/5: Work on Case 2.1

· Th 10/7: Complete Case 2.1, start ICE 4

· Tu 10/12: ICE 4, Start Case 2.2

· Th 10/14: Case 2.2 Class Optional: Work on Case 2.2 Remotely. I’m available.

· Tu 10/19: Case 3.1

· Th 10/21: Finish Case 2.2, Start Case 3.1