Customer data present a host of complex statistical challenges, from atypical distributions and rare events, to censoring and panel mortality, to ambiguous causal pathways. This seminar empowers participants to leverage the latest innovations in data science to address these issues and uncover actionable customer insights. The seminar is a balance of theory and practice: The statistical and methodological issues for each topic are carefully explained at an intuitive level, and then participants run through guided, hands-on examples using R and Python.

There are no formal prerequisites for this course, but participants should have a basic familiarity with statistical concepts, such as regression and inference. For participants without exposure to R and Python (e.g., if a participant is not sure what a dataframe is), we can suggest short tutorials to bring them up to speed prior to attending the seminar.

Topics

Brand Loyalty: NPS and Competitive Loyalty Studies

Insights from Behavioral Economics: Anchoring, Confirmation Bias, Conjoint Analysis

Optimization through Experimental Design and A/B Testing

Propensity Scoring Methods

Cluster Analysis & Principal Components

Special Statistical Issues: Censoring

Special Statistical Issues: Rare Event Analysis

Special Statistical Issues: Time to Event Analysis

Intro to Sentiment Analysis & NLP

Schedule

Day 1, morning session 1: Brand Loyalty

Statistical Challenges/Solutions for NPS and Competitive Loyalty Studies

Day 1, morning session 2: Insights from Behavioral Economics

Anchoring, Confirmation Bias, Conjoint Analysis

Day 1, afternoon session 1: Cloud-Based Development Environment

Review student applications and data

AWS console

Security and user permissions

Manage EC2 instances & S3 buckets

Amazon SageMaker

Day 1, afternoon session 2: Practice & Review with Cloud-Based Tools

Practice with Python, Jupyter, git, and AWS training data tools

Day 2, morning session 1: Experimental Design

A/B Testing and Beyond

Day 2, morning session 2: Propensity Scoring Methods

Propensity Scoring: Methodology, Pitfalls, and Discussion

Day 2, afternoon session 1: Practice & Review

Apply A/B testing and propensity scoring

Day 2, afternoon session 2: Clustering and Dimensional Analysis

Principal Components, Factor Analysis, K-Means Clustering

Day 3, morning session 1: Practice & Review

Implement principal Components, factor analysis, and k-means using student or instructor-supplied data

Day 3, morning session 2: Special Statistical Issues – Censoring

Censoring in Customer Analytics: Special Topics & Solutions

Day 3, afternoon session 1: Special Statistical Issues – Rare Event Analysis

Rare Event Analysis: Special Topics & Solutions

Day 3, afternoon session 2: Special Statistical Issues – Time to Event Analysis

Time to Event Analysis: Special Topics & Solutions

Day 4, morning session 1: Introduction to Sentiment Analysis & NLP

Leveraging the Power of Natural Language Processing: An Introduction

Day 4, morning session 2:

Review specific topics as a class or small groups

Guided development and problem solving

1.1 development coaching

Day 4, afternoon session:

Optional individualized 1.1 assistance

Ensure each student leaves with a functioning application