<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=204513679968251&amp;ev=PageView&amp;noscript=1">

Do You Even Data

A data-driven marketing blog

Want to learn how you can translate incredible data list information into killer marketing campaigns? Want to better understand how data research and models can enhance the data you already have?
All Posts

Change in Likelihood of Recommendation According to Nonlinear Model

A Comparison between Linear and Nonlinear Models

For decades, market researchers have relied on linear regression models to perform key driver analyses. These models are designed to reveal how a predictor, such as the use of particular product, may determine an outcome, such as the likelihood that a customer becomes a brand advocate who recommends the company’s products to their friends. Some sophisticated models even include interactions that show how the relationship between a predictor and an outcome vary over some critical feature such as a customer’s age.

Change in Likelihood of Recommendation According to Linear Model

The problem with this classic approach is that the relationship between the predictor and the outcome is assumed to be linear. As you will see soon, this is something that should not be taken at face value. Customer behavior is not always that simple and linear graphs can over simplify the reality of the situation at hand. By using a nonlinear model[1],[2] it is possible to gain a more accurate picture of how the likelihood of recommending a brand varies by age.

 Change in Likelihood of Recommendation According to Nonlinear Model-

A comparison between the two models reveals how linear key driver analyses may conceal valuable information. Nonlinear models help you hear the voice of the customer more clearly. They can be used to examine such issues as whether customer purchase patterns change after an important life event or whether today’s customers are behaving differently than the customers of previous quarters.

Model Comparison

The potential for nonlinear models is vast, but their application requires experience and technical know how. Find out more about how nonlinear models can help your business.


Learn More 



[1] Hastie T, Tibshirani R. Varying-coefficient models. Journal of the Royal Statistical Society, Series B. 1993; 55:757–779.
[2] Hoover DR, Rice JA, Wu CO, Yang LP. Nonparametric smoothing estimates of time-varying coefficient models with longitudinal data. Biometrika. 1998; 85:809–822.