Predictive Modeling

Predictive modeling encompasses a variety of techniques from statistics and data mining that analyze current and historical data to make predictions about future events.

Apple’s Taylor Swift Marketing Campaign Analysis

Elections eBook

Michael Lieberman's practical ebook, Giving You The Edge - The Science of Winning Elections, identifies and explains the use of key research methodology and multivariate analysis in supporting political campaign goals through the various stages of an election, including a summary chart matching objectives, analytical techniques and results.

The Quirk’s Marketing Research Review article, Data Fusion – How Researchers Can Create C – Suite Deliverables, discusses interpretations and presentation of predictive analytics and marketing research, which both employ data scientists.

A cruise line example illustrates how to use discrete choice to determine marginal value in this piece published in Quirk’s Marketing Research Review, How to Price an Island.

Boosting Employee Retention with Predictive Analytics discusses how to increase business efficiency by using models to identify those employees most at risk of leaving a company.

Predictive Regression Analysis explains how to employ the widely used and robust predictive technique made famous in Ian Ayres’ seminal work, SuperCrunchers.

Regression Analysis Positioning uses the most common predictive analytic method to distinguish between two types of customer and maximizes how to communicate to each.

Finding Supporters – Political Predictive Analytics uses a regression technique to target potential supporters for candidate Liz Anderson’s bid for Congress.

Donor Predictive CHAID Tree uses a classification tree method to box in most-likely donors to a political party.

Campaigns and Elections – What is Driving the Vote? describes the successful employment of regression analysis to determine voter motives.

The article How Nate Silver Did It looks at how probabilistic models, which use weighted averages to predict election outcomes, can be used in state and local races as well as in presidential ones.

Measuring and Using Employee Satisfaction, a piece that illustrates ‘Within human resources the dependent variable should not only be defined by the HR goals of the employer … It should also include an evaluation of the impact on customer satisfaction,’ outlines a data-driven method to predict which attitudes will lower employee turnover.

Predictive modeling encompasses a variety of techniques from statistics and data mining that analyze current and historical data to make predictions about future events.

In business, predictive models exploit patterns found in historical and transactional data to identify risks and opportunities. Models capture relationships among many factors to allow assessment of risk or potential associated with a particular set of conditions, guiding decision making for candidate transactions.

Predictive analytics adds great value to a businesses decision making capabilities by allowing it to formulate smart policies on the basis of predictions of future outcomes.

Multivariate Solutions offers broad range of tools and techniques for this type of analysis. Their selection is determined by the analytical sophistication of the firm as well as the nature of the problem being solved.

Multivariate Solutions offers three main types of predictive analytics:

  • Predictive models
  • Descriptive models
  • Decision models

Predictive Analytics is often applied to any of the following projects:

  • Analytical Customer Relationship Management (CRM):
  • Data Mining
  • Direct marketing
  • Customer retention
  • Collection analytics
  • Product or economy level prediction
  • Sales or penetration forecasting
  • Loyalty Analytics
  • Predicative Modeling
  • Voter Identification