Abstract:In this paper, which consists of two parts (Part I: representation results; Part II: estimation and forecasting methods), we set up the theoretical foundations for a high‐dimensional functional factor model approach in the analysis of large cross‐sections (panels) of functional time series (FTS). In Part I, we establish a representation result stating that, under mild assumptions on the covariance operator of the cross‐section, we can represent each FTS as the sum of a common component driven by scalar factors loaded via functional loadings, and a mildly cross‐correlated idiosyncratic component. Our model and theory are developed in a general Hilbert space setting that allows for mixed panels of functional and scalar time series. We then turn, in Part II, to the identification of the number of factors, and the estimation of the factors, their loadings, and the common components. We provide a family of information criteria for identifying the number of factors, and prove their consistency. We provide average error bounds for the estimators of the factors, loadings, and common components; our results encompass the scalar case, for which they reproduce and extend, under weaker conditions, well‐established similar results. Under slightly stronger assumptions, we also provide uniform bounds for the estimators of factors, loadings, and common components, thus extending existing scalar results. Our consistency results in the asymptotic regime where the number N of series and the number T of time observations diverge thus extend to the functional context the "blessing of dimensionality" that explains the success of factor models in the analysis of high‐dimensional (scalar) time series. We provide numerical illustrations that corroborate the convergence rates predicted by the theory, and provide a finer understanding of the interplay between N and T for estimation purposes. We conclude with an application to forecasting mortality curves, where we demonstrate that our approach outperforms existing methods

Forecasting High-Dimensional Functional Time Series: Application to Sub-National Age-Specific Mortality

High-dimensional functional time series forecasting: An application to age-specific mortality rates

FORECASTING MULTIPLE FUNCTIONAL TIME SERIES IN A GROUP STRUCTURE: AN APPLICATION TO MORTALITY

Grouped Functional Time Series Forecasting: An Application to Age-Specific Mortality Rates

Dynamic principal component regression for forecasting functional time series in a group structure

Grouped multivariate and functional time series forecasting: An application to annuity pricing

Robust forecasting of mortality and fertility rates: A functional data approach

A Forecast-driven Hierarchical Factor Model with Application to Mortality Data

Mortality and life expectancy forecasting for a group of populations in developed countries: A multilevel functional data method

On the modelling and prediction of high-dimensional functional time series

Forecasting density-valued functional panel data

Nonparametric Trend Estimation in Functional Time Series with Application to Annual Mortality Rates

Multipopulation mortality modelling and forecasting: The multivariate functional principal component with time weightings approaches

Mortality and life expectancy forecasting for a group of populations in developed countries: A robust multilevel functional data method

DYNAMIC PRINCIPAL COMPONENT REGRESSION: APPLICATION TO AGE-SPECIFIC MORTALITY FORECASTING

Factor Models for High‐Dimensional Functional Time Series II: Estimation and Forecasting

Mortality Forecasting using Factor Models: Time-varying or Time-invariant Factor Loadings?

A Bayesian Functional Data Model for Surveys Collected under Informative Sampling with Application to Mortality Estimation using NHANES

A New Fourier Approach under the Lee-Carter Model for Incorporating Time-Varying Age Patterns of Structural Changes

Reconciling Forecasts of Infant Mortality Rates at National and Sub-National Levels: Grouped Time-Series Methods

Factor Models for High‐Dimensional Functional Time Series I: Representation Results