Statistical Modelling 14 (4) (2014), 315–335

Functional convolution models

Maria Asencio
Mosaic ATM


Giles Hooker
Department of Biological Statistics and Computational Biology,
Cornell University
e-mail: gjh27@cornell.edu


H Oliver Gao
Department of Civil and Environmental Engineering,
Cornell University


Abstract:

This article considers the application of functional data analysis methods to modelling particulate matter emission profiles from dynamometer experiments. In particular the functional convolution model is introduced as an extension of the distributed lag model to functional (smooth and continuous) observations. We present a penalized ordinary least squares estimator for the model and a novel bootstrap procedure to provide pointwise confidence regions for the estimated convolution functions. The model is illustrated on the Coordinating Research Council E55/59 study of diesel truck emissions.

Keywords:

Functional data analysis; time series; smoothing; convolution; historical linear model

Downloads:

Example data in zipped archive
back