Statistics and Modeling for Direct Marketers Seminar Outline
Day 1 | Day 2
Day 1: 9:00 A.M. – 5:00 P.M.
Please arrive 30 minutes earlier than the start time to register.
Basic Statistical Review and Applications
- How “chance” affects the results from DM list and package tests
- How to adjust for these “chance” effects (normal distributions etc.)
- How to build forecasts for the results from your next promotion
- How to determine your needed sample size
- How to test if two list or package results are “statistically different”
Predictive Modeling and Segmentation Modeling
- What the difference is between the two types of modeling procedures, how they relate to each other, and when to use each
Correlation Analysis
- How to read scatter diagrams and plots of residuals
- How to read a correlation matrix, and what it tells you to avoid
Multiple Regression Analysis
- When to use a multiple regression program
- How to analyze data prior to regression analysis and why it’s imperative to perform this step
- How to model the non-linear response patterns so common in direct marketing response analyses
- How to use “dummy variables” in regression analysis
- How to perform simple regression analyses using Excel™
AID/CHAID or "Tree" Analysis
- How to interpret the output of an AID/CHAID program
- Why an AID/CHAID analysis is not used for scoring
- How AID/CHAID can be used to choose variables for a regression model
Day 2: 9:00 A.M. – 4:00 P.M.
Factor Analysis
- When to use factor analysis prior to using regression
- How to interpret the output of a factor analysis
- How to use factor analysis to analyze your marketing research data
Cluster Analysis
- How to use a cluster analysis to create your own customer target market groups
- How to decide on the “right” number of clusters to include in a cluster analysis
Discriminant Analysis
- The similarities and the differences between discriminant analysis and regression analysis. When to use each.
Logistic Regression
- What it is
- Why it is the recommended method to be used when performing response analysis
- How it is different from ordinary regression and/or discriminant analysis
How to Get Started
- What hardware and software you need
- How to determine if you should use consultants, service bureau, or do it in-house
- What you can expect to gain in the first year
* Outline is subject to change.
