×

Forecasting functional time series. (English) Zbl 1293.62267

Summary: We propose forecasting functional time series using weighted functional principal component regression and weighted functional partial least squares regression. These approaches allow for smooth functions, assign higher weights to more recent data, and provide a modeling scheme that is easily adapted to allow for constraints and other information. We illustrate our approaches using age-specific French female mortality rates from 1816 to 2006 and age-specific Australian fertility rates from 1921 to 2006, and show that these weighted methods improve forecast accuracy in comparison to their unweighted counterparts. We also propose two new bootstrap methods to construct prediction intervals, and evaluate and compare their empirical coverage probabilities.

MSC:

62P25 Applications of statistics to social sciences
62H25 Factor analysis and principal components; correspondence analysis
62G08 Nonparametric regression and quantile regression
62G09 Nonparametric statistical resampling methods
62J07 Ridge regression; shrinkage estimators (Lasso)
91D20 Mathematical geography and demography
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Aguilera, A. M.; Escabias, M.; Valderrama, M. J., Using principal components for estimating logistic regression with high-dimensional multicollinear data, Computational Statistics & Data Analysis, 50, 8, 1905-1924 (2006) · Zbl 1445.62190
[2] Alho, J. M., Remarks on the use of probabilities in demography and forecasting, (Keilman, N., Perspectives on mortality forecasting (2005), Swedish Social Insurance Agency: Swedish Social Insurance Agency Oslo), 27-38
[3] Alho, J. M.; Spencer, B. D., Statistical demography and forecasting (2005), Springer: Springer New York · Zbl 1080.62103
[4] Bellman, R., Adaptive control processes: A guided tour (1961), Princeton University Press · Zbl 0103.12901
[5] Cai, T.; Hall, P., Prediction in functional linear regression, Annals of Statistics, 34, 5, 2159-2179 (2006) · Zbl 1106.62036
[6] Chatfield, C., Calculating interval forecasts, Journal of Business & Economic Statistics, 11, 2, 121-135 (1993)
[7] Chatfield, C., Time series forecasting (2000), Chapman & Hall/CRC: Chapman & Hall/CRC Boca Raton
[8] Dauxois, J.; Pousse, A.; Romain, Y., Asymptotic theory for the principal component analysis of a vector random function: Some applications to statistical inference, Journal of Multivariate Analysis, 12, 1, 136-154 (1982) · Zbl 0539.62064
[9] Delaigle, A., Hall, P., & Apanasovich, T. V. (2009). Weighted least squares methods for prediction in the functional data linear model. arXiv:0902.3319v1http://arxiv.org/abs/0902.3319v1; Delaigle, A., Hall, P., & Apanasovich, T. V. (2009). Weighted least squares methods for prediction in the functional data linear model. arXiv:0902.3319v1http://arxiv.org/abs/0902.3319v1 · Zbl 1326.62151
[10] Erbas, B.; Hyndman, R. J.; Gertig, D. M., Forecasting age-specific breast cancer mortality using functional data models, Statistics in Medicine, 26, 2, 458-470 (2007)
[11] Escabias, M.; Aguilera, A. M.; Valderrama, M. J., Principal component estimation of function logistic regression: Discussion of two different approaches, Journal of Nonparametric Statistics, 16, 3-4, 365-384 (2004) · Zbl 1065.62114
[12] Faber, N., Uncertainty estimation for multivariate regression coefficients, Chemometrics and Intelligent Laboratory Systems, 64, 2, 169-179 (2002)
[13] Fernández Pierna, J. A.; Jin, L.; Wahl, F.; Faber, N. M.; Massart, D. L., Estimation of partial least squares regression prediction uncertainty when the reference values carry a sizeable measurement error, Chemometrics and Intelligent Laboratory Systems, 65, 2, 281-291 (2003)
[14] Greenshtein, E., Best subset selection, persistence in high-dimensional statistical learning and optimization under \(l_1\) constraint, Annals of Statistics, 34, 5, 2367-2386 (2006) · Zbl 1106.62022
[15] Greenshtein, E.; Ritov, Y., Persistence in high-dimensional linear predictor selection and the virtue of overparameterization, Bernoulli, 10, 6, 971-988 (2004) · Zbl 1055.62078
[16] Hall, P.; Horowitz, J., Methodology and convergence rates for functional linear regression, Annals of Statistics, 35, 1, 70-91 (2007) · Zbl 1114.62048
[17] Hall, P.; Hosseini-Nasab, M., On properties of functional principal components analysis, Journal of the Royal Statistical Society: Series B, 68, 1, 109-126 (2006) · Zbl 1141.62048
[18] Hall, P.; Hosseini-Nasab, M., Theory for high-order bounds in functional principal components analysis, Mathematical Proceedings of the Cambridge Philosophical Society, 146, 1, 225-256 (2009) · Zbl 1153.62050
[19] Hall, P.; Müller, H.; Wang, J., Properties of principal component methods for functional and longitudinal data analysis, Annals of Statistics, 34, 3, 1493-1517 (2006) · Zbl 1113.62073
[20] Human Mortality Database, (2008). University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany). Data downloaded on 10 Sep 2008 www.mortality.org; Human Mortality Database, (2008). University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany). Data downloaded on 10 Sep 2008 www.mortality.org
[21] Hyndman, R. J. (2007). addb: Australian Demographic Data Bank. R package version 3.222. http://www.robhyndman.info/Rlibrary/addb; Hyndman, R. J. (2007). addb: Australian Demographic Data Bank. R package version 3.222. http://www.robhyndman.info/Rlibrary/addb
[22] Hyndman, R. J.; Booth, H., Stochastic population forecasts using functional data models for mortality, fertility and migration, International Journal of Forecasting, 24, 3, 323-342 (2008)
[23] Hyndman, R. J.; Koehler, A. B.; Ord, J. K.; Snyder, R. D., Forecasting with exponential smoothing: The state space approach (2008), Springer: Springer Berlin · Zbl 1211.62165
[24] Hyndman, R. J., & Shang, H. L. (2008a). ftsa: Functional time series analysis. R package version 1.0. http://monashforecasting.com/index.php?title=R_packages; Hyndman, R. J., & Shang, H. L. (2008a). ftsa: Functional time series analysis. R package version 1.0. http://monashforecasting.com/index.php?title=R_packages
[25] Hyndman, R. J., & Shang, H. L. (2008b). Rainbow plots, bagplots, and boxplots for functional data. Working paper 9/08, Department of Econometrics & Business Statistics, Monash University. http://www.buseco.monash.edu.au/depts/ebs/pubs/wpapers/2008/9-08.php; Hyndman, R. J., & Shang, H. L. (2008b). Rainbow plots, bagplots, and boxplots for functional data. Working paper 9/08, Department of Econometrics & Business Statistics, Monash University. http://www.buseco.monash.edu.au/depts/ebs/pubs/wpapers/2008/9-08.php
[26] Hyndman, R. J.; Ullah, M. S., Robust forecasting of mortality and fertility rates: A functional data approach, Computational Statistics & Data Analysis, 51, 10, 4942-4956 (2007) · Zbl 1162.62434
[27] Kargin, V.; Onatski, A., Curve forecasting by functional autoregression, Journal of Multivariate Analysis, 99, 10, 2508-2526 (2008) · Zbl 1151.62073
[28] Krämer, N.; Boulesteix, A.; Tutz, G., Penalized partial least squares with applications to \(B\)-spline transformations and functional data, Chemometrics and Intelligent Laboratory Systems, 94, 1, 60-69 (2008)
[29] Martens, H.; Naes, T., Multivariate calibration (1989), Wiley: Wiley Chichester · Zbl 0732.62109
[30] Mesle, F.; Vallin, J., Reconstitution of annual life tables for nineteenth-century France, Population: An English Selection, 3, 33-62 (1991)
[31] Ng, P. T., & Maechler, M. (2008). COBs: COBS — Constrained \(B\) http://wiki.r-project.org/rwiki/doku.php?id=packages:cran:cobs; Ng, P. T., & Maechler, M. (2008). COBs: COBS — Constrained \(B\) http://wiki.r-project.org/rwiki/doku.php?id=packages:cran:cobs
[32] Preda, C.; Saporta, G., Clusterwise PLS regression on a stochastic process, Computational Statistics & Data Analysis, 49, 1, 99-108 (2005) · Zbl 1429.62299
[33] Preda, C.; Saporta, G., PLS regression on a stochastic process, Computational Statistics & Data Analysis, 48, 1, 149-158 (2005) · Zbl 1429.62224
[34] Ramsay, J. O.; Dalzell, C. J., Some tools for functional data analysis (with discussion), Journal of the Royal Statistical Society: Series B, 53, 3, 539-572 (1991) · Zbl 0800.62314
[35] Ramsay, J. O.; Silverman, B. W., Applied functional data analysis (2002), Springer: Springer New York · Zbl 1011.62002
[36] Ramsay, J. O.; Silverman, B. W., Functional data analysis (2005), Springer: Springer New York · Zbl 1079.62006
[37] Reiss, P. T.; Ogden, T. R., Functional principal component regression and functional partial least squares, Journal of the American Statistical Association, 102, 479, 984-996 (2007) · Zbl 1469.62237
[38] Rice, J. A.; Silverman, B. W., Estimating the mean and covariance structure nonparametrically when the data are curves, Journal of the Royal Statistical Society: Series B, 53, 1, 233-243 (1991) · Zbl 0800.62214
[39] Silverman, B. W., Incorporating parametric effects into functional principal components analysis, Journal of the Royal Statistical Society: Series B, 57, 4, 673-689 (1995) · Zbl 0827.62051
[40] Silverman, B. W., Smoothed functional principal components analysis by choice of norm, Annals of Statistics, 24, 1, 1-24 (1996) · Zbl 0853.62044
[41] Wedderburn, R. W.M., Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method, Biometrika, 61, 3, 439-447 (1974) · Zbl 0292.62050
[42] Wilmoth, J.; Andreev, K. F.; Jdanov, D. A.; Glei, D. A., Methods protocol for the human mortality database (2007)
[43] Wold, H., Soft modelling by latent variables: The non-linear iterative partial least squares (NIPALS) approach, (Gani, J., Perspectives in probability and statistics (1975), Applied Probability Trust: Applied Probability Trust Sheffield), 117-145 · Zbl 0331.62058
[44] Wood, S. N., Monotonic smoothing splines fitted by cross validation, SIAM Journal on Scientific Computing, 15, 5, 1126-1133 (1994) · Zbl 0821.65002
[45] Yao, F.; Müller, H. G.; Wang, J. L., Functional linear regression analysis for longitudinal data, Annals of Statistics, 33, 6, 2873-2903 (2005) · Zbl 1084.62096
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.