Kion Kim.

Ph. D. Graduate student.

Recent history functional linear models (RHFLM)

Abstract:

We propose a variant of historical functional linear models for cases where the current response is affected by the predictor process in a window into the past. Different from the rectangular support of functional linear models, the triangular support of the historical functional linear models and the point-wise support of the varying coefficient models, the current model has a sliding window support into the past.

This idea leads to models that bridge the gap between varying coefficient models and functional linear (historic) models. We propose an algorithm for this model that can be applied to longitudinal data where the measurements are taken on irregular time points and missing values are allowed.

The proposed estimation algorithm is shown to be fast, involving one dimensional basis expansions and one dimensional smoothing procedures.

Authors: Kion Kim, Damla Senturk, Runzi Li