DA 302W: Predictive Analytics

Textbook Information

Primary Book:

Principles of Business Forecasting, 2nd Edition 

Author: Keith Ord, Robert Fildes, and Nikolaos Kourentzes 

ISBN: 978-0999064917 

 

Secondary Book:

Data Mining and Predictive Analytics, 2nd Edition 

Author: Daniel T. Larose, and Chantal D. Larose 

ISBN: 978-1118116197 

 

*You are encouraged to purchase both books. The secondary book is used to a lesser extent in the course.

Published Remarks

None

Hardware Requirements

None

Software Requirements

None

Proctored Exams

None

Course Description

DA 302W is a four-credit course with lecture and writing components. (DA201W, DA 302W, and DA 401W each carry only one credit of "writing"; completion of all three courses is required to meet the writing requirement.) The principal objective of this course is to help students predict “what will happen” in future events based on historical data. This course exposes students to predictive analytics techniques that are consistent with best practices in the data analytics industry. Students will learn how to properly examine problem contexts to use the most appropriate method to develop the best predictive model. Students will also learn how to evaluate their results and interpret findings to users at different levels of an organization. This course focuses on the application of spreadsheets, scripting languages for data analytics, and current statistical software packages.

Course Learning Objectives:

By the end of this course, successful students will be able to: 

  • Explain the fundamental concepts, importance, and applications of predictive analytics 
  • Utilize basic techniques to assess data quality and conduct data processing  
  • Use appropriate measures to evaluate the performance of predictive models 
  • Implement time series analysis methods in predictive analytics 
  • Construct and interpret linear regression models to explore relationships between variables and predict outcomes 
  • Explain the basic concepts of machine learning 
  • Implement machine learning techniques to classify data, identify patterns, and predict outcomes 
  • Employ tools (such as spreadsheet modelin and R) to perform predictive analyses 
  • Present predictive insights through written reports and visualizations 
  • Derive communication about predictive results and insights to diverse audiences 

 General Overview of assignments:

Quizzes (about 10%): Test your understanding of the core concepts and methods covered in the lessons. 

Homework Assignments (about 40%): Designed to let you use the methods and models introduced in the lessons to solve problems. Some of the assignments are case-based, which require you to identify/define the problem(s), select technique(s) to solve, and create report to summarize findings and insights. 

Discussion (about 10%): Designed to let you engage with classmates and collaboratively explore predictive analytics topics. 

Midterm Exam (20%): Test your understanding of the key knowledge and your ability to apply the core techniques and methods learned in the first half of the course. 

Final Paper (20%): The final paper is a comprehensive project where you will identify a problem critical to improving performance or achieving a specific goal. You will use various predictive analytics tools and techniques to analyze the problem, generate prediction, and develop conclusions, insights, and communicate findings through a report.