Generalized Additive Models.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Monotonic Calibrated Interpolated Look-Up Tables (2015, background): “terpretable function class in that the parameters dictate the relative importance of each feature. Linear approaches can be generalized to sum nonlinear components, as in generalized additive models (Hastie and Tibshirani, 1990) and some kernel methods, while still retaining some of their interpreta”
MatchIt: Nonparametric Preprocessing for Parametric Causal Inference (2011, methodology): “Generalized additive distance measures are implemented via the mcgv package (Hastie and Tibshirani 1990).”
Causal Inference With General Treatment Regimes (2004, methodology): “In parallel with the aforementioned twopart model, we use the generalized additive model (Hastie and Tibshirani 1990) with the binomial family and logistic link to model the probability of positive medical costs, and use the Gaussian family and identity link to model the conditional distribution of ”