The Burke Institute presents
 

Practical Multivariate Analysis On-Site Training

In this seminar, totally devoid of matrix notations or derivations, you will learn how to select and interpret the best multivariate technique for analyzing the complex relationships between many variables typically encountered in marketing research studies.

Course Description/Agenda

WHAT YOU WILL LEARN

• What multivariate techniques are, what they do and when to use them.
• How to choose between Regression, Discriminant, Factor, Cluster, M.D.S., Correspondence, Conjoint, DCM, CHAID, Logit, Canonical and other techniques to answer specific management questions from your research data.
• How to interpret multivariate results and communicate what they reveal to management.
• Classic and contemporary approaches to analytical modeling
• Machine learning / AI Techniques
• Bayesian Network Analysis

WORKSHOP AGENDA

SESSION 1: CLASSIC & CONTEMPORARY APPROACHES TO ANALYTICAL MODELING

• Univariate, bivariate, multivariate and unstructured data analytical techniques overview
• Comparing traditional and new ways of analyzing data
• Tools for helping to link and analyze multiple data sources together
• Review of fundamental principles for analyzing data

SESSION 2: SELECTING THE MOST APPROPRIATE ANALYTICAL TECHNIQUES

• A structured framework of seven charts to help select the best analytical techniques for specific marketing research applications
• Exercise Selecting the most appropriate analytical techniques for several applications
• A series of case examples applying a wide range of data analysis techniques.

SESSION 3: MULTIPLE REGRESSION ANALYSIS

• Applications of regression
• Interpretation of computer regression output
• Understanding and dealing with multicollinearity
• Stepwise regression
• Data transformations, modeling interactions, and other improvements to the regression model

SESSION 4: DISCRIMINANT, DISCRETE CHOICE, AND CONJOINT ANALYSIS

• Discriminant analysis description and applied examples
• What are conjoint and discrete choice analysis
• Case study illustrating the use of conjoint and discrete choice
• Analyzing conjoint and discrete choice results
• Simulation example

SESSION 5: FACTOR AND CLUSTER ANALYSIS

• Factor and cluster analysis: What are they, how they work and when they should be used
• Performing PCA/factor analysis: inputs and outputs
• How Principal Components Analysis differs from exploratory factor analysis
• Hierarchical cluster analysis and K-means cluster analysis
• Other forms of cluster analysis including LCA
• Exercise using factor and cluster analysis

SESSION 6: MACHINE LEARNING / AI TECHNIQUES

• Overview of machine learning and AI
• Dynamic learning models vs static learning models
• Four main pillars of AI and ML
• The relationship between artificial intelligence, machine learning and deep learning
• Using machine learning and AI techniques for text analysis

SESSION 7: BAYESIAN NETWORK ANALYSIS (BNA)

• Overview of Bayesian analysis
• Case studies using Bayesian analysis
• Instructor lead demonstration leveraging Bayesian analysis
• Practical applications of Bayesian analysis

SESSION 8: CASE STUDY (GROUP WORKSHOP): ANALYSIS PLANS & OVERALL CUSTOMER LOYALTY MEASUREMENT

• Overview of case study exercise
• Review survey information and data table results
• Review supporting charts and graphs
• Determine what additional analysis techniques to apply to the data
• Provide recommendations to help inform business decisions

More Seminar Information

The Burke Institute Delivery Method  On-Site Training