Abstract
The paper provides a kind of Klimko–Nelson’s theorems alternative in the case of conditional least-squares and M-estimators for array time series when the assumptions of almost sure convergence cannot be established. We do not assume stationarity nor even local stationarity. Besides, we provide sufficient conditions for two of the assumptions and a procedure for the evaluation of the information matrix in array time series. In addition to time-dependent models, illustrations to a threshold model and a count data model are given.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Abramowitz M, Stegun IA (1965) Handbook of mathematical functions with formulas, graphs, and mathematical tables. Dover Publications, New York
Alj A, Azrak R, Mélard G (2014) On conditions in central limit theorems for martingale difference arrays. Econ Lett 123:305–307
Alj A, Azrak R, Ley C, Mélard G (2017) Asymptotic properties of QML estimators for VARMA models with time-dependent coefficients. Scand J Stat 44:617–635
Alj A, Azrak R, Mélard G (2021) General estimation results for td-VARMA models. Submitted
Amendola A, Niglio M, Vitale C (2007) The autocorrelation functions in SETARMA models. In: Kontoghiorghes EJ, Gatu C (eds) Optimisation, econometric and financial analysis. Springer, Berlin, pp 127–142
Azrak R, Mélard G (2006) Asymptotic properties of quasi- likelihood estimators for ARMA models with time-dependent coefficients. Stat Infer Stoch Process 9:279–330
Bai ZD, Wu Y (1997) General M-estimation. J Multivar Anal 63:117–135
Bardet J-M, Doukhan P (2018) Non-parametric estimation of time varying AR(1)-processes with local stationarity and periodicity. Electron J Stat 12:2323–2354
Bardet J-M, Doukhan P, Wintenberger O (2018) Contrast estimation of general locally stationary processes using coupling. arXiv preprint arXiv:2005.07397
Basawa IV, Prakasa Rao BLS (1980) Statistical inference for stochastic processes. Academic Press, New York
Brown BM (1971) Martingale central limit theorems. Ann Math Stat 42:59–66
Brown BM, Eagleson GK (1971) Martingale convergence to infinitely divisible laws with finite variances. Trans Am Math Soc 162:449–453
Chen R, Tsay RS (1993) Functional-coefficient autoregressive models. J Am Stat Assoc 88:298–308
Dahlhaus R (2012) Locally stationary processes. In: Subba Rao T, Subba Rao S, Rao CR (eds) Handbook of statistics. Time series analysis: methods and applications, vol 30. Elsevier, Amsterdam, pp 351–413
Davidson JEH (1994) Stochastic limit theorems. An introduction for econometricians. Oxford University Press, Oxford
Davis RA, Dunsmuir WTM, Streett SB (2003) Observation driven models for Poisson counts. Biometrika 90:777–790
Doukhan P, Fokianos K, Tjøstheim D (2012) On weak dependence conditions for Poisson autoregressions. Stat Probab Lett 82:942–948
Erdélyi A (1953) Higher Transcendental Functions, vol 1. McGraw-Hill, New York
Francq C, Zakoïan J-M (1998) Estimating linear representations of nonlinear processes. J Stat Plan Inference 68:145–165
Fokianos K, Støve B, Tjøstheim D, Doukhan P (2020) Multivariate count autoregression. Bernoulli 26(1):471–499
Gaenssler P, Strobel J, Stute W (1978) On central limit theorems for martingale arrays. Acta Mathematica Academiae Scientiarum Hungaricae 31:205–216
Giordano F, Niglio M, Vitale CD (2013) Ergodicity of threshold autoregressive models: the two regimes case. https://www.sis-statistica.it/old_upload/contenuti/2013/09/RS12-On-the-stationarity-of-the-Threshold.pdf
Hall P, Heyde CC (1980) Martingale limit theory and its application. Academic Press, New York
Han F, Wu W (2019) Probability inequalities for high dimensional time series under a triangular array framework. arXiv preprint arXiv:1907.06577
Klimko LA, Nelson PI (1978) On conditional least squares estimation for stochastic processes. Ann Stat 6:629–642
Lai TL (1994) Asymptotic properties of nonlinear least squares estimates in stochastic regression models. Ann Stat 22:1917–1930
Lehmann EL, Casella G (1998) Theory of point estimation. Springer, New York
Liboschik T, Fokianos K, Fried R (2017) tscount: an R package for analysis of count time series following generalized linear models. J Stat Softw 82(5)
Martin RD (1981) Robust methods for time series. In: Findley DF (ed) Applied time series II. Academic Press, New York, pp 683–759
Mélard G (2021) An indirect proof for the asymptotic properties of VARMA model estimators. Econ Stat (in press). https://doi.org/10.1016/j.ecosta.2020.12.004
Mélard G, Niglio M (2021) Another approach for asymptotic properties of threshold models. In preparation
Muler N, Peña D, Yohai V (2009) Robust estimation for ARMA models. Ann Stat 37:816–840
Muma M, Zoubir AM (2017) Bounded influence propagation \(\tau \)-estimation: a new robust method for ARMA model estimation. IEEE Trans Signal Process 65(7):1712–1727
Nelson PI (1980) A note on strong consistency of least squares estimators in regression models with martingale difference errors. Ann Stat 8:1057–1064
Petruccelli JD, Woolford SW (1984) A threshold AR(1) model. J Appl Probab 21:270–286
Roueff F, Sánchez-Pérez A (2018) Prediction of weakly locally stationary processes by auto-regression. ALEA Latin Am J Probab Math Stat 15:1215–1239
Taniguchi M, Kakizawa Y (2000) Asymptotic theory of statistical inference for time series. Springer, New York
Teräsvirta T (1994) Specification, estimation, and evaluation of smooth transition autoregressive models. J Am Stat Assoc 89:208–218
Tjøstheim D (1986) Estimation in nonlinear time series models. Stoch Process Appl 21:251–273
Tjøstheim D (1994) Non-linear time series: a selective review. Scand J Stat 21:97–130
Tong H (1990) Nonlinear time series: a dynamical systems approach. Oxford University Press, Oxford
Tong H, Lim KS (1980) Threshold autoregression, limit cycles and cyclical data. J Roy Stat Soc B 42:245–292
Yin J (2019) A note on mixing in high dimensional time series. arXiv preprint arXiv:1911.10648
Zoubir AM, Koivunen V, Ollila E, Muma M (2018) Robust statistics for signal processing. Cambridge University Press, Cambridge
Acknowledgements
Guy Mélard has benefited from a Belgian research grant F.R.S.-FNRS 1.5.261.09. We thank Prof. Nikolaos Kourogenis, University of Piraeus (Greece), for providing the reference of Brown and Eagleson (1971). We thank also Abdelkamel Alj for his comments on the proof of Theorem 2.1. The second author thanks Marcella Niglio (University of Salerno, Italy) and Luisa Bisaglia (University of Padua, Italy) for their invitations to their universities, Marcella Niglio for her remarks, and Konstantinos Fokianos for references and discussions on count data models. This paper was presented at several places by the second author who thanks Aboubacar Amiri, Marianne Clausel, Christian Francq, Marc Hallin, Yacouba Boubacar Mainassara, and Davy Paindaveine, among others. Finally, we are grateful to the Editor and to the two referees (in particular, we appreciated the insistence of one of them), whose suggestions lead to a much better paper.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Azrak, R., Mélard, G. Asymptotic properties of conditional least-squares estimators for array time series. Stat Inference Stoch Process 24, 525–547 (2021). https://doi.org/10.1007/s11203-021-09242-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11203-021-09242-8
Keywords
- Klimko–Nelson’s theorems
- Non-stationary process
- Multivariate time series
- Time-varying models
- Information matrix