Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
144 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Forecasting density-valued functional panel data (2403.13340v2)

Published 20 Mar 2024 in stat.ME

Abstract: We introduce a statistical method for modeling and forecasting functional panel data represented by multiple densities. Density functions are nonnegative and have a constrained integral and thus do not constitute a linear vector space. We implement a center log-ratio transformation to transform densities into unconstrained functions. These functions exhibit cross-sectional correlation and temporal dependence. Via a functional analysis of variance decomposition, we decompose the unconstrained functional panel data into a deterministic trend component and a time-varying residual component. To produce forecasts for the time-varying component, a functional time series forecasting method, based on the estimation of the long-run covariance, is implemented. By combining the forecasts of the time-varying residual component with the deterministic trend component, we obtain $h$-step-ahead forecast curves for multiple populations. Illustrated by age- and sex-specific life-table death counts in the United States, we apply our proposed method to generate forecasts of the life-table death counts for 51 states.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (79)
  1. Drewnowski’s index to measure lifespan variation: Revisiting the Gini coefficient of the life table. Theoretical Population Biology 148, 1–10.
  2. Ahn, S. C. and A. R. Horenstein (2013). Eigenvalue ratio test for the number of factors. Econometrica 81(3), 1203–1227.
  3. Aitchison, J. (1982). The statistical analysis of compositional data. Journal of the Royal Statistical Society. Series B (Methodological) 44(2), 139–177.
  4. Thirty years on: A review of the lee–carter method for forecasting mortality. International Journal of Forecasting 39(3), 1033–1049.
  5. An age-at-death distribution approach to forecast cohort mortality. Insurance: Mathematics and Economics 91, 129–143.
  6. An age-at-death distribution approach to forecast cohort mortality. Insurance: Mathematics and Economics 91(C), 129–143.
  7. Coherent forecasts of mortality with compositional data analysis. Demographic Research 37, 527–566.
  8. Booth, H. (2006). Demographic forecasting: 1980 to 2005 in review. International Journal of Forecasting 22(3), 547–581.
  9. Mortality modelling and forecasting: A review of methods. Annals of Actuarial Science 3(1-2), 3–43.
  10. Measuring the longevity risk in mortality projections. Bulletin of the Swiss Association of Actuaries 2, 105–130.
  11. Canudas-Romo, V. (2010). Three measures of longevity: Time trends and record values. Demography 47(2), 299–312.
  12. An autocovariance-based learning framework for high-dimensional functional time series. Journal of Econometrics 239(2), 105385.
  13. Nonstationarity in time series of state densities. Journal of Econometrics 192(1), 152–167.
  14. Chatfield, C. (1993). Calculating interval forecasts. Journal of Business & Economic Statistics 11(2), 121–135.
  15. Chatfield, C. (2000). Time-Series Forecasting. Boca Raon: CRC Press.
  16. Chiou, J.-M. (2012). Dynamical functional prediction and classification with application to traffic flow prediction. The Annals of Applied Statistics 6(4), 1588–1614.
  17. Condino, F. (2023). Share density-based clustering of income data. Statistical Analysis and Data Mining 16, 336–347.
  18. A high dimensional functional time series approach to evolution outlier detection for grouped smart meters. Quality Engineering 35(3), 371–387.
  19. Emerson, J. D. and D. C. Hoaglin (1983). Analysis of two-way tables by medians. In D. C. Hoaglin, F. Mosteller, and J. W. Tukey (Eds.), Understanding Robust and Exploratory Data Analysis, pp.  166–210. New York: Wiley.
  20. Finite sample theory for high-dimensional functional/scalar time series with applications. Electronic Journal of Statistics 16(1), 527–591.
  21. Principal component analysis for compositional data with outliers. Environmetrics 20(6), 621–632.
  22. Jensen-Shannon divergence and Hilbert space embedding. In International Symposium on Information Theory, pp.  30.
  23. High-dimensional functional time series forecasting: An application to age-specific mortality rates. Journal of Multivariate Analysis 170, 232–243.
  24. Probabilistic forecasting. The Annual Review of Statistics and Its Application 1, 125–151.
  25. Gneiting, T. and A. E. Raftery (2007). Strictly proper scoring rules, prediction, and estimation. Journal of the American Statistical Association: Review Article 102(477), 359–378.
  26. On consistency and sparsity for high-dimensional functional time series with application to autoregressions. Bernoulli 29(1), 451–472.
  27. Factor modelling for high-dimensional functional time series. Working paper, arXiv.
  28. Assessing the finite dimensionality of functional data. Journal of the Royal Statistical Society: Series B 68(4), 689–705.
  29. Factor models for high-dimensional functional time series I: Representation results. Journal of Time Series Analysis 44(5-6), 578–600.
  30. Dynamics of financial returns densities: A functional approach applied to the bovespa intraday index. International Journal of Forecasting 34, 75–88.
  31. Adaptive bandwidth selection in the long run covariance estimator of functional time series. Computational Statistics and Data Analysis 100, 676–693.
  32. Simplicial principal component analysis for density functions in Bayes space. Computational Statistics & Data Analysis 94, 330–350.
  33. Human Mortality Database (2024). Max Planck Institute for Demographic Research (Germany), University of California, Berkeley (USA), and French Institute for Demographic Studies (France). Available at www.mortality.org. (data downloaded on June 14, 2023).
  34. forecast: Forecasting functions for time series and linear models. Monash University. R package version 8.21.1. URL: https://CRAN.R-project.org/package=forecast.
  35. Coherent mortality forecasting: the product-ratio method with functional time series models. Demography 50(1), 261–283.
  36. Automatic time series forecasting: The forecast package for r. Journal of Statistical Software 27(3), 1–22.
  37. Hyndman, R. J. and H. L. Shang (2009). Forecasting functional time series. Journal of the Korean Statistical Society 38(3), 199–211.
  38. Foreacsting the old-age dependency ratio to determine a sustainable pension age. Australian and New Zealand Journal of Statistics 63(2), 241–256.
  39. Forecasting high-dimensional functional time series: Application to sub-national age-specific mortality. Journal of Computational and Graphical Statistics in press.
  40. Kneip, A. and K. J. Utikal (2001). Inference for density families using functional principal component analysis. Journal of the American Statistical Association: Theory and Methods 96(454), 519–532.
  41. Forecasting of density functions with an application to cross-sectional and intraday returns. International Journal of Forecasting 35(4), 1304–1317.
  42. Kullback, S. and R. A. Leibler (1951). On information and sufficiency. The Annals of Mathematical Statistics 22(1), 79–86.
  43. Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root? Journal of Econometrics 54(1), 159–178.
  44. Detection and estimation of structural breaks in high-dimensional functional time series. Technical report, The University of York.
  45. Long-range dependent curve time series. Journal of the American Statistical Association: Theory and Methods 115, 957–971.
  46. Local Whittle estimation of long-range dependence for functional time series. Journal of Time Series Analysis 42(5-6), 685–695.
  47. Mercer, J. and A. R. Forsyth (1909). Xvi. functions of positive and negative type, and their connection the theory of integral equations. Philosophical Transactions of the Royal Society of London. Series A, Containing Papers of a Mathematical or Physical Character 209(441-458), 415–446.
  48. Mosteller, F. and J. W. Tukey (1977). Data Analysis and Regression. Cambridge: Addison-Wesley.
  49. Oeppen, J. (2008). Coherent forecasting of multiple-decrement life tables: A test using Japanese cause of death data. In European Population Conference. European Association for Population Studies.
  50. Paparoditis, E. and H. L. Shang (2023). Bootstrap prediction bands for functional time series. Journal of the American Statistical Association: Theory and Methods 118(542), 972–986.
  51. The maximum entropy mortality model: forecasting mortality using statistical moments. Scandinavian Actuarial Journal 2019(8), 661–685.
  52. Pollard, J. H. (1987). Projection of age-specific mortality rates. In Population Bulletin of the United Nations, 21/22, pp. 55–69. Population Commission.
  53. Functional Data Analysis. Springer Series in Statistics. New York: Springer.
  54. Rice, G. and H. L. Shang (2017). A plug-in bandwidth selection procedure for long-run covariance estimation with stationary functional time series. Journal of Time Series Analysis 38(4), 591–609.
  55. Robine, J.-M. (2001). Redefining the stages of the epidemiological transition by a study of the dispersion of life spans: the case of France. Population: An English Selection 13(1), 173–193.
  56. Robust principal component analysis for power transformed compositional data. Journal of the American Statistical Association: Theory and Methods 110(509), 136–148.
  57. Scealy, J. L. and A. H. Welsh (2017). A directional mixed effects model for compositional expenditure data. Journal of the American Statistical Association: Applications and Case Studies 112(517), 24–36.
  58. Seo, W.-K. and B. K. Beare (2019). Cointegrated linear processes in Bayes Hilbert space. Statistics & Probability Letters 147, 90–95.
  59. Shang, H. L. (2019). Dynamic principal component regression: Application to age-specific mortality forecasting. ASTIN Bulletin 49(3), 619–645.
  60. Forecasting age distribution of death counts: An application to annuity pricing. Annals of Actuarial Science 14(1), 150–169.
  61. Multi-population modelling and forecasting life-table death counts. Insurance: Mathematics and Economics 106, 239–253.
  62. Shang, H. L. and R. J. Hyndman (2017). Grouped functional time series forecasting: An application to age-specific mortality rates. Journal of Computational and Graphical Statistics 26, 330–343.
  63. A multilevel functional data method for forecasting population, with an application to the United Kingdom. International Journal of Forecasting 32(3), 629–649.
  64. Forecasting Australian subnational age-specific mortality rates. Journal of Population Research 38(1), 1–24.
  65. Shannon, C. E. (1948). A mathematical theory of communication. Bell System Technical Journal 27(3), 379–423.
  66. Sun, Y. and M. G. Genton (2012). Functional median polish. Journal of Agricultural, Biological, and Environmental Statistics 17(3), 354–376.
  67. Graphical principal component analysis of multivariate functional time series. Journal of the American Statistical Association: Theory and Methods in press.
  68. Clustering and forecasting multiple functional time series. The Annals of Applied Statistics 16(4), 2523–2553.
  69. Factor models for high‐dimensional functional time series II: Estimation and forecasting. Journal of Time Series Analysis 44(5-6), 601–621.
  70. Tukey, J. (1977). Exploratory Data Analysis. Reading: Addison-Wesley.
  71. United States Mortality Database (2024). University of California, Berkeley (USA). Department of Demography at the University of California, Berkeley. Available at usa.mortality.org (data downloaded on March 15, 2023).
  72. van Raalte, A. A. and H. Caswell (2013). Perturbation analysis of indices of lifespan variability. Demography 50(5), 1615–1640.
  73. Lifespan variation by occupational class: Compression or stagnation over time? Demography 51, 73–95.
  74. Life expectancy and disparity: An international comparison of life table data. BMJ Open 1(1), e000128.
  75. Functional data analysis for sparse longitudinal data. Journal of the American Statistical Association: Theory and Methods 100(470), 577–590.
  76. Zhang, J.-T. (2014). Analysis of Variance for Functional Data. Boca Raon: Chapman & Hall/CRC.
  77. Functional linear regression of cumulative distribution functions. Working paper, arXiv.
  78. Statistical inference for high-dimensional panel functional time series. Journal of the Royal Statistical Society Series B: Statistical Methodology 85(2), 523–549.
  79. Modeling Financial Time Series with S-PLUS. New York: Springer.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets