Explore the vast range of titles available.

We study a functional version of fractionally integrated time series that covers the nonstationary case when the memory parameter d is above 0.5. We project time series, with varying levels of nonstationarity, onto a finite-dimensional subspace. We obtain the eigenvalues and eigenfunctions that span a sample version of the dominant subspace through dynamic functional principal component analysis of the sample long-run covariance functions. Within the context of functional autoregressive fractionally integrated moving average models, we evaluate and compare finite-sample bias and mean-squared error among some time-and frequency-domain Hurst exponent estimators via Monte Carlo simulations. We apply the estimators to Canadian female and male life-table death counts. Les auteurs de cette étude explorent une version fonctionnelle de séries chronologiques stationnaires à intégration d’ordre fractionnaire. Comme la non stationnarité est présente lorsque le paramètre de mémoire d est supérieur à 0.5, les auteurs ont pu faire varier ce paramètre pour effectuer des projections de séries chronologiques à stationnarités variables sur un sous-espace de dimension finie. Cet exercice a permis d’obtenir des valeurs propres et fonctions propres qui ont servi à recouvrer une version échantillonnale du sous-espace dominant et ce grâce à une analyse dynamique en composantes principales fonctionnelles appliquée aux fonctions de covariance empirique à long terme. En simulant des modèles fonctionnels de moyennes mobiles autorégressives à intégration d’ordre fractionnaire, les auteurs ont évalué et comparé les biais et les erreurs quadratiques moyennes d’estimateurs de l’exposant de Hurst évalués dans les domaines temporel et fréquentiel. Une application à des données réelles relatives à la table de mortalité des femmes et des hommes au Canada est présentée.

We introduce methods and theory for functional or curve time series with long-range dependence. The temporal sum of the curve process is shown to be asymptotically normally distributed, the conditions for this covering a functional version of fractionally integrated autoregressive moving averages. We also construct an estimate of the long-run covariance function, which we use, via functional principal component analysis, in estimating the orthonormal functions spanning the dominant subspace of the curves. In a semiparametric context, we propose an estimate of the memory parameter and establish its consistency. A Monte Carlo study of finite-sample performance is included, along with two empirical applications. The first of these finds a degree of stability and persistence in intraday stock returns. The second finds similarity in the extent of long memory in incremental age-specific fertility rates across some developed nations. Supplementary materials for this article are available online.

Accuracy in fertility forecasting has proved challenging and warrants renewed attention. One way to improve accuracy is to combine the strengths of a set of existing models through model averaging. The model-averaged forecast is derived using empirical model weights that optimise forecast accuracy at each forecast horizon based on historical data. We apply model averaging to fertility forecasting for the first time, using data for 17 countries and six models. Four model-averaging methods are compared: frequentist, Bayesian, model confidence set, and equal weights. We compute individual-model and model-averaged point and interval forecasts at horizons of one to 20 years. We demonstrate gains in average accuracy of 4–23% for point forecasts and 3–24% for interval forecasts, with greater gains from the frequentist and equal weights approaches at longer horizons. Data for England and Wales are used to illustrate model averaging in forecasting age-specific fertility to 2036. The advantages and further potential of model averaging for fertility forecasting are discussed. As the accuracy of model-averaged forecasts depends on the accuracy of the individual models, there is ongoing need to develop better models of fertility for use in forecasting and model averaging. We conclude that model averaging holds considerable promise for the improvement of fertility forecasting in a systematic way using existing models and warrants further investigation.

As a forward-looking measure of future equity market volatility, the VIX index has gained immense popularity in recent years to become a key measure of risk for market analysts and academics. We consider discrete reported intraday VIX tick values as realisations of a collection of curves observed sequentially on equally spaced and dense grids over time and utilise functional data analysis techniques to produce 1-day-ahead forecasts of these curves. The proposed method facilitates the investigation of dynamic changes in the index over very short time intervals as showcased using the 15-s high-frequency VIX index values. With the help of dynamic updating techniques, our point and interval forecasts are shown to enjoy improved accuracy over conventional time series models.

Functional data analysis tools, such as function-on-function regression models, have received considerable attention in various scientific fields because of their observed high-dimensional and complex data structures. Several statistical procedures, including least squares, maximum likelihood, and maximum penalized likelihood, have been proposed to estimate such function-on-function regression models. However, these estimation techniques produce unstable estimates in the case of degenerate functional data or are computationally intensive. To overcome these issues, we proposed a partial least squares approach to estimate the model parameters in the function-on-function regression model. In the proposed method, the B-spline basis functions are utilized to convert discretely observed data into their functional forms. Generalized cross-validation is used to control the degrees of roughness. The finite-sample performance of the proposed method was evaluated using several Monte-Carlo simulations and an empirical data analysis. The results reveal that the proposed method competes favorably with existing estimation techniques and some other available function-on-function regression models, with significantly shorter computational time.

Visualization methods help in the discovery of characteristics that might not have been apparent by using mathematical models and summary statistics. However, visualization methods have not received much attention in demography, with the exceptions of scatter plots and Lexis surfaces. We utilize a phase plane plot to visualize the rate of change, obtained from derivatives of a continuous function. The phase plane plot bears a resemblance to hysteresis loops, isogrowth curves and solutions to differential equations. Using Australian and Chilean fertility, we present phase plane plots. Similarly to the scatter plot and Lexis surface, the phase plane plot identifies the age with maximum fertility rate and displays skewness of the fertility distribution. Unlike the scatter plot and Lexis surface, the phase plane plot identifies the age with maximum positive or negative velocity (i.e. the trend) and can compare the magnitude of the rate of change between any two years on the basis of the size of the radius of circles. The phase plane plot enables visualization of dynamic changes in fertility for a given age over the years and is potentially useful for visualizing dynamic changes in birth cohort fertility. Via the animate package in LATEX, a dynamic phase plane plot is also proposed to visualize changes in fertility over age or year.

A key summary statistic in a stationary functional time series is the long-run covariance function that measures serial dependence. It can be consistently estimated via a kernel sandwich estimator, which is the core of dynamic functional principal component regression for forecasting functional time series. To measure the uncertainty of the long-run covariance estimation, we consider sieve and functional autoregressive (FAR) bootstrap methods to generate pseudo-functional time series and study variability associated with the long-run covariance. The sieve bootstrap method is nonparametric (i.e., model-free), while the FAR bootstrap method is semi-parametric. The sieve bootstrap method relies on functional principal component analysis to decompose a functional time series into a set of estimated functional principal components and their associated scores. The scores can be bootstrapped via a vector autoregressive representation. The bootstrapped functional time series are obtained by multiplying the bootstrapped scores by the estimated functional principal components. The FAR bootstrap method relies on the FAR of order 1 to model the conditional mean of a functional time series, while residual functions can be bootstrapped via independent and identically distributed resampling. Through a series of Monte Carlo simulations, we evaluate and compare the finite-sample accuracy between the sieve and FAR bootstrap methods for quantifying the estimation uncertainty of the long-run covariance of a stationary functional time series.

A partial least squares regression is proposed for estimating the function-on-function regression model where a functional response and multiple functional predictors consist of random curves with quadratic and interaction effects. The direct estimation of a function-on-function regression model is usually an ill-posed problem. To overcome this difficulty, in practice, the functional data that belong to the infinite-dimensional space are generally projected into a finite-dimensional space of basis functions. The function-on-function regression model is converted to a multivariate regression model of the basis expansion coefficients. In the estimation phase of the proposed method, the functional variables are approximated by a finite-dimensional basis function expansion method. We show that the partial least squares regression constructed via a functional response, multiple functional predictors, and quadratic/interaction terms of the functional predictors is equivalent to the partial least squares regression constructed using basis expansions of functional variables. From the partial least squares regression of the basis expansions of functional variables, we provide an explicit formula for the partial least squares estimate of the coefficient function of the function-on-function regression model. Because the true forms of the models are generally unspecified, we propose a forward procedure for model selection. The finite sample performance of the proposed method is examined using several Monte Carlo experiments and two empirical data analyses, and the results were found to compare favorably with an existing method.

In areas of application, including actuarial science and demography, it is increasingly common to consider a time series of curves; an example of this is age-specific mortality rates observed over a period of years. Given that age can be treated as a discrete or continuous variable, a dimension reduction technique, such as principal component analysis (PCA), is often implemented. However, in the presence of moderate-to-strong temporal dependence, static PCA commonly used for analyzing independent and identically distributed data may not be adequate. As an alternative, we consider a dynamic principal component approach to model temporal dependence in a time series of curves. Inspired by Brillinger’s (1974, Time Series: Data Analysis and Theory. New York: Holt, Rinehart and Winston) theory of dynamic principal components, we introduce a dynamic PCA, which is based on eigen decomposition of estimated long-run covariance. Through a series of empirical applications, we demonstrate the potential improvement of 1-year-ahead point and interval forecast accuracies that the dynamic principal component regression entails when compared with the static counterpart.

In this study, we propose a function-on-function linear quantile regression model that allows for more than one functional predictor to establish a more flexible and robust approach. The proposed model is first transformed into a finitedimensional space via the functional principal component analysis paradigm in the estimation phase. It is then approximated using the estimated functional principal component functions, and the estimated parameter of the quantile regression model is constructed based on the principal component scores. In addition, we propose a Bayesian information criterion to determine the optimum number of truncation constants used in the functional principal component decomposition. Moreover, a stepwise forward procedure and the Bayesian information criterion are used to determine the significant predictors for including in the model. We employ a nonparametric bootstrap procedure to construct prediction intervals for the response functions. The finite sample performance of the proposed method is evaluated via several Monte Carlo experiments and an empirical data example, and the results produced by the proposed method are compared with the ones from existing models.