A0169
Title: Sparse high-dimensional and functional autoregressions
Authors: Xinghao Qiao - London School of Economics (United Kingdom) [presenting]
Shaojun Guo - Renmin University of China (China)
Abstract: Multivariate functional data arise in a broad spectrum of real applications. However, many studies in high dimensional functional data focus primarily on the critical assumption of independent and identically distributed (i.i.d.) samples. We consider a sparse high dimensional functional autoregressive model to characterize the dynamic dependence across different functional time series. We propose a new regularization method via group lasso to estimate the autoregressive coefficient functions and derive non-asymptotic bounds for the estimation errors of the regularized estimates. We also introduce a measure for stationary functional processes that provides insight into the effect of dependence on the accuracy of the regularized estimates. Finally, we show that the proposed model significantly outperforms its competitors through both simulated and real data sets.
