BUS 655 - Business Forecasting

One of the most important roles of a person in business is to make better decisions. In BUS 550 you began to understand the decision process and how to incorporate information. BUS 655 expands on that basic structure in two major ways.

New method of modeling data 

  1. Adaptive and non-linear models (regression +) 
  2. General non-linear methods (Synthetic Neural Networks-SNN) 
  3. Time series methods

Development of an automated structure to support decisions tied to data - Decision Support Systems (DSS)

In BUS 550 we developed models to predict the expected dollars contributed by a town and then use that information to decide which towns to canvas first. This is a problem that is analogous to many business situations that follow the path of:

  1. need to make a decision; 
  2. collect data; 
  3. develop model; 
  4. predict values; 
  5. make decision; 
  6. repeat often or do only once.

This course is a very "hands on" working with data course, working with sample data sets or even better bringing problems from work. Through sharing of experience and discussion of MANY data sets and problems we gain years of experience in a few months. Grading is based on 4 projects and a presentation that, normally group work. The course is structured to challenge the very good quantitative people while providing a path to success for the numerically challenged. Along the way, subjects we will cover include

  1. Decision Support Systems; 
  2. Collecting data; 
  3. Data sources; 
  4. Data-Bases ( as tools ); 
  5. Review regression; 
  6. Expanded regression; 
  7. Time dependent models and analysis; 
  8. Advanced Solver; 
  9. Time series techniques, 
  10. Exponential smoothing, 
  11. Moving averages, 
  12. Box-Jenkins ( ARIMA); 
  13. Synthetic Neural Networks; 
  14. Chaos Theory; 
  15. Genetic Algorithms

Examples I will cover will include: Financial models - Stock prices, Risk, including bond ratings, Cash flow; Behavior models - Customer attrition, Customer likes/dislikes; Buying patterns - Propensity to buy; Politics - Identify swing voters; and Sales.

We will also work on any data set that student choose to share with the class and get help on. We often yield instantly applicable results for students at work. We will also maintain privacy as needed by your company.

~~~

One of the most important rolls of a person in business is to make better decisions. In this course, you will begin to understand the decision process and how to incorporate information.  BUS 655P expands on that basic structure in two major ways.

  1. New method of modeling data
  2. Adaptive and non-linear models (regression +)
  3. General non-linear methods (Synthetic Neural Networks-SNN) Time series methods
  4. Development of a automated structure to support decisions tied to data Decision Support Systems (DSS)

In this course, we developed models to predict the expected dollars contributed by a town and then use that information to decide which towns to canvas first.  This is a problem that is analogous to many business situations that follow the path of:  need to make a decision, collect data, develop model, predict values, make decision, repeat often or do only once.  This course is a very “hands on” working with data course, working with sample data sets or even better bringing problems from work.  Through sharing of experience and discussion of MANY data sets and problems we gain years of experience in a few months.  Grading is based on 4 projects and a presentation that, normally group work.  The course is structured to challenge the very good quant people while providing a path to success for the numerically challenged. 

Cross-Listed

  • BUS 655P