Regression analysis
-
Factorization Machines for High Cardinality Features (Part 4 of 4)
This is the fourth in a 4-part series where Anders Larson and Shea Parkes discuss predictive analytics with high cardinality features. In the prior episodes we focused on approaches to handling individual high cardinality features, but these methods did not explicitly address feature interactions. Factorization Machines can responsibly estimate all pairwise interactions, even when multiple high cardinality features are included. With a healthy selection of high cardinality features, a well tuned Factorization Machine can produce results that are more accurate than any other learning algorithm. -
The Hashing Trick for High Cardinality Features (Part 3 of 4)
This is the third in a 4-part series where Anders Larson and Shea Parkes discuss predictive analytics with high cardinality features. In this episode they focus on feature engineering via the hashing trick. The hashing trick is most applicable for extremely high cardinality, and at first glance can seem almost ridiculous. In a lot of ways, it is the same as bucketing values at random. But there are times that it is more valuable to include randomly engineered buckets than to exclude the original high cardinality feature entirely. -
Predictive Analytics Hack-a-Thon 2022
Describes the Predictive Analytics Hack-a-Thon held in 2022 and its outcomes. -
Y-aware Feature Engineering with High Cardinality Features (Part 2 of 4)
This is the second in a 4-part series where Anders Larson and Shea Parkes discuss predictive analytics with high cardinality features. In this episode they focus on y-aware feature engineering. Y-aware feature engineering is all about carefully bleeding information from your training response back into your engineered features without grossly misrepresenting your ability to generalize to new data. -
Introducing Predictive Analytics with High Cardinality Features (Part 1 of 4)
This is the first in a 4-part series where Anders Larson and Shea Parkes discuss predictive analytics with high cardinality features. In this episode they focus on introducing what high cardinality features are, why you might want to work with them and some of the basic approaches to including them in predictive models. -
Anders vs. Shea, Part 4: A Champion is Crowned
Shea Parkes, FSA, MAAA, and Anders Larson, FSA, MAAA, reveal the results of the competition and share some final thoughts on the 2021 Milliman Health Practice Hackathon. -
Anders vs. Shea, Part 2: Anders’ Story
Shea Parkes, FSA, MAAA, and Anders Larson, FSA, MAAA, are joined by Nick Vander Heyden to discuss the approach used by Anders’ team in the 2021 Milliman Health Practice Hackathon. -
Emerging Topics Community: Anders vs. Shea, Part 1: Setting the Stage
Shea Parkes, FSA, MAAA, and Anders Larson, FSA, MAAA, are are joined by the organizers of the 2021 Milliman Health Practice Hackathon: Riley Heckel, FSA, MAAA, Austin Barrington, FSA, MAAA, and Phil Ellenberg. -
Using a Statistical Framework and AI Techniques to Enhance Basic Actuarial Assumptions Part 2: Application to Accelerated Underwriting
The paper uses the actuarial and statistical framework developed in Part 1 of the paper and discusses an application of this model in accelerated underwriting. It is a methodical way to quantify the impact of various risk factors in the underwriting process besides age and gender and provide an actuarially rigorous process to evaluate the aggregate risk level of a new policyholder. -
Surrogate Models: A Comfortable Middle Ground?
In this article we examine the benefits and drawbacks of using a global surrogate of a black-box model. Are surrogates a comfortable middle ground that provide insurers high predictive power as well as model explainability?
Welcome to our medium.
Dive into insightful content. Gain practical knowledge. Explore the latest research and key information for future use. Discover the Emerging Topics Community, an online forum that focuses on three main topic areas–Modeling, Predictive Analytics and Futurism, and Technology.