BU425 Business Analytics Fall 2022
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
BU425 Business Analytics Fall 2022
Core content for quiz/midterm
General Information
There will be a 2.5 hour in class quiz (via MyLS quizzes) on Wednesday, November 16th, 2021 from 7:00PM – 9:30PM. It will feature multiple choice and short answer questions (including caselet analysis) covering the core content of the course described in this document. You should be familiar with the case analysis performed throughout the semester.
Students who know in advance of the exam that they will not be attending must confirm via e-mail prior to the start of the exam. All students who do not write the quiz will have to take a differed quiz (covering the entire course content) during the week after the last scheduled class (week of Dec12th- 16th).
Quiz is going to be administered online through MyLS without “LockDown Browser” and it will be open book. However, as we are not going to use lockdown browser the integrity of the exam is at your discretion. You are not permitted to use internet or any communication with your computer/tablet/phone. Please read the academic misconduct policy of Wilfrid Laurier University.
You will not be coding in R.
Core Content
Questions related to coding in R will be only from the material covered in power point “S1 - R Tutorial.pptx” . However, you should be familiar with the analysis of all cases uploaded to MyLS!
Linear regression
• Regression modeling with multiple dependent variables
o Regression model and estimation of coefficients
o Transforming the dependent variable
o Adding additional features: higher order polynomial terms and interaction terms
o Dummy variables
• Regression assumptions incl. multicollinearity and impact of outliers.
• Interpreting regression output
o 2 , p-value of variables
• Business applications of linear regression (incl. prediction)
Logistic regression
• When to use logistic vs. linear regression
• Model selection
• Interpreting a logistic regression
o Interpret Fi and p-value
• Business applications
Classification and predictive analytics
• What is a classification problem and how is it set up
• k-nearest neighbours, tree induction, support vector machines, neural networks
o intuition
o selecting between models
o complexity
• Logistic regression:
o How logistic regression can be used as a classifier
o Linear classifier and adjusting classification threshold
o Model selection
o Embedded methods to reduce complexity: regularization
• Desired qualities of a classifier
• Classification error
o training and testing error
o different types of error (false positives and false negatives)
• Understand generalizability
o overfitting, model complexity
• Business applications.
o decision making
o expected value and confusion matrix
o recommendation systems
▪ user and item based collaborative filtering
• Neural networks
Unsupervised learning
• KNN
• Principal component analysis
• Clustering
PLUS, all the course content (including cases and tutorials) covered up to and including week 8 (November 9!).
2022-11-14