Functional linear regression that's interpretable

TL;DR

This paper introduces FLiRTI, a new functional linear regression method that produces interpretable coefficient functions with exact zeros in regions of no relationship, using variable selection on derivatives, supported by theoretical error bounds.

Contribution

The paper presents FLiRTI, a novel approach combining variable selection and derivative analysis to produce interpretable, sparse, and accurate coefficient functions in functional linear regression.

Findings

FLiRTI effectively identifies zero regions in b2(t) for interpretability.

The method achieves accurate estimation with simple, interpretable structures.

Theoretical bounds support the robustness of FLiRTI.

Abstract

Regression models to relate a scalar $Y$ to a functional predictor $X(t)$ are becoming increasingly common. Work in this area has concentrated on estimating a coefficient function, $\beta(t)$, with $Y$ related to $X(t)$ through $\int\beta(t)X(t) dt$. Regions where $\beta(t)\ne0$ correspond to places where there is a relationship between $X(t)$ and $Y$. Alternatively, points where $\beta(t)=0$ indicate no relationship. Hence, for interpretation purposes, it is desirable for a regression procedure to be capable of producing estimates of $\beta(t)$ that are exactly zero over regions with no apparent relationship and have simple structures over the remaining regions. Unfortunately, most fitting procedures result in an estimate for $\beta(t)$ that is rarely exactly zero and has unnatural wiggles making the curve hard to interpret. In this article we introduce a new approach which uses variable selection ideas, applied to various derivatives of $\beta(t)$, to produce estimates that are both interpretable, flexible and accurate. We call our method "Functional Linear Regression That's Interpretable" (FLiRTI) and demonstrate it on simulated and real-world data sets. In addition, non-asymptotic theoretical bounds on the estimation error are presented. The bounds provide strong theoretical motivation for our approach.