Model Development, Validation, & Evaluation


Overview/Description
Target Audience
Prerequisites
Expected Duration
Lesson Objectives
Course Number
Expertise Level



Overview/Description
Analytic model management ensures that models are not only superior to alternatives, but they also meet or exceed current business needs. Explore the process of building, validating, and evaluating a predictive analytics model.

Target Audience
All individuals who are new to predictive analytics and wish to use it to optimize their business performance; business leaders; analysts; marketing, sales, software, and IT professionals who want to add predictive analytics to their skill set; and decision makers of any kind

Prerequisites
None

Expected Duration (hours)
1.0

Lesson Objectives

Model Development, Validation, & Evaluation

  • start the course
  • recognize key phases in developing a model
  • recognize important aspects of data preparation for model development
  • identify key functions in the model creation process
  • identify key considerations of model validation
  • balance model complexity with overfitting
  • identify features of two-fold validation
  • recognize key features of Occam's Razor
  • calculate mean squared error measures
  • recognize prediction variation measures
  • recognize evaluation measures for classification models
  • interpret lift and gain charts
  • interpret ROC curves and AUC
  • determine the superior classification model
  • Course Number:
    df_prma_a18_it_enus

    Expertise Level
    Intermediate