Member-only story
Optimizing Prices with PyGAM
A Smooth Approach to Revenue Maximization
Price optimization is challenging and important. Companies often struggle to set their prices. Even giants like Disney have that problem.
Prices set too low can create a perception of poor quality, while high prices can discourage people from buying. That’s where Generalized Additive Models (GAMs) come in, offering a flexible and powerful alternative.
PyGAM is a neat Python library that makes implementing GAMs much easier.
Let’s dive in!
Understanding Generalized Additive Models (GAMs)
GAMs are a flexible version of linear models. They do not assume a straight-line relationship between your variables and the outcome, but instead, GAMs let you model those relationships with smoothed lines (or curves). The total effect will be the sum of individual effects for each factor affecting the price.
This flexibility is achieved through “smoothness” using splines to capture complex patterns without going overboard and overfitting your data. Compared to other models, GAMs are awesome because they’re interpretable (you can see how each factor impacts the price) and have built-in regularization to prevent overfitting.
