Title: Prediction using averaging estimated functional linear regression models
Authors: Xinyu Zhang - Academy of Mathematics and Systems Science, Chinese Academy of Sciences (China)
Jeng-Min Chiou - Academia Sinica (Taiwan) [presenting]
Yanyuan Ma - Pennsylvania State University (United States)
Abstract: A novel model averaging approach is proposed to predict the functional response variable. We develop a cross-validation model averaging estimator based on functional linear regression models in which the response and the covariate are both treated as random functions. The weights chosen by the method are asymptotically optimal in the sense that the squared error loss of the predicted function is as small as that of the infeasible best possible averaged function. Monte Carlo studies and a data application indicate that in most cases the approach performs better than model selection.