Predictive modeling

Prerequisites:

• Basic computer science principles and skills
• Basics in data analysis

Courses:

• Advanced Data Analysis&Big Data for Business Intelligence
• Data Analysis

Predictive modeling is focused on the creation and using models for prediction the probability of an output in accordance with a signal, given by a set amount of input data. Predictive modeling can be used in many business cases, such as customer relationship management, risk management, process management. The course deals with connections between business objectives of the predictive models and use of the models. The course gives understanding of process of building predictive model, appropriate tools and software, explains how to monitor and maintain predictive models and plan for future model updating. Course special focus is on the data for predictive modeling. Many cases of predictive models from different areas will be considered.

Topics include:

• Key parts of predictive models
• Predictive modeling process
• Data-preprocessing
• Over-fitting and model tuning
• Regression model. Case studies
• Classification models. Case studies
• Other models. Case studies
• Measuring predictor importance
• Feature selection
• Factors that can affect model performance
• Updating predictive model
• Predictive modeling software and tools
• Prediction in customer relationship management

1. Eric Siegel, Thomas H. Davenport. Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die. Wiley, 1 edition, 2013
2. Thomas W. Miller. Modeling Techniques in Predictive Analytics: Business Problems and Solutions with R. FT Press, 1 edition, 2013.
3. Conrad Carlberg.  Predictive Analytics: Microsoft Excel. Que Publishing, 1 edition, 2012.
4. Max Kuhn,  Kjell Johnson. Applied Predictive Modeling. Springer, 2013.
5. Foster Provost, Tom Fawcett.  Data Science for Business: What you need to know about data mining and data-analytic thinking. O'Reilly Media, 1 edition, 2013.