Theory of Probability and Mathematical Statistics
Spectral estimation in the presence of missing data
Natalia Bahamonde, Paul Doukhan
Download PDF
Abstract: In this article we propose a quasi-Whittle estimator for parametric families of time series models in the presence of missing data. This estimator extends results to the incompletely observed case. This extension is valid to non-Gaussian and non-linear models. It also allows to bound the variance of an associated quasiperiodogramm. A simulation study validates empirically the proposed estimate for mixing and non-mixing models.
Keywords: Limit theorems, Time Series, Auto Correlation, Whittle estimator, Weakly dependent
Bibliography: 1.Andrews, D. Non strong mixing autoregressive processes. J. Appl. Probab., 21:930–934 (1984).
2. Bardet, J., Doukhan, P., and Leo´n, J. Uniform limit theorems for the periodogram of weakly dependent time series and their applications to whittle’s estimate. J. of Time Series Anal., 29(1):906–945 (2008).
3. Bondon, P. and Bahamonde, N. Least squares estimation of arch models with missing observations. Journal of Time Series Analysis, 33 (6):880–891 (2012).
4. Dahlhaus, R. A likelihood approximation for locally stationary processes. Annals of Statistics, 28:1762–1794. (2000).
5. Dedecker, J. and Doukhan, P. A new covariance inequality and applications. Stochastic Process. Appl., 106(1):63–80 (2003).
6. Doukhan, P. Mixing: Properties and examples, volume 85 of Lecture Notes in Statistics. Springer-Verlag, New York. (1994).
7. Doukhan, P. and Lang, G. Evaluation for moments of a ratio with application to regression estimation. Bernoulli, 15(4):1259–1286 (2009).
8. Doukhan, P. and Leo´n, J. R. Cumulants for stationary mixing random sequences and applications to empirical spectral density. Probab. Math. Statist., 10(1):11–26 (1989).
9. Doukhan, P. and Louhichi, S. A new weak dependence condition and applications to moment inequalities. Stochastic Process. Appl., 84(2):313–342 (1999).
10. Doukhan, P., Massart, P., and Rio, E. The functional central limit theorem for strongly mixing processes. Ann. Inst. H. Poincar´e Probab. Statist., 30(1):63–82 (1994).
11. Doukhan, P., Mayo, N., and Truquet, T. Weak dependence, models and some applications. Metrika, 69, 199–225 (2009).
12. Doukhan, P., Teyssi`ere, G., and Winant, P. A LARCH(∞) vector valued process. In Dependence in probability and statistics, volume 187 of Lecture Notes in Statist., pages 245–258. Springer, New York (2006).
13. Dunsmuir, W. and Robinson, P. M. Asymptotic theory for time series containing missing and amplitude modulated observations. Sankhy¯a Ser. A, 43(3):260–281 (1981a).
14. Dunsmuir, W. and Robinson, P. M. Estimation of time series models in the presence of missing data. J. Amer. Statist. Assoc., 76(375):560–568 (1981b).
15. Efromovich, S. Efficient non-parametric estimation of the spectral density in presence of missing observations. Journal of Time Series Analysis, 35:407–427 (2014).
16. Engle, R. F. Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. Econometrica, 50(4):987–1007.
17. Giraitis, L. and Robinson, P. M. (2001). Whittle estimation of arch models. Econometric Theory, 17(3):608–631 (1982).
18. Knight, M. I., Nunes, M. A., and Nason, G. P. Spectral estimation for locally stationary time series with missing observations. Stat. Comput., 22(4):877–895 (2012).
19. Lavielle, M. Detection of multiple changes in a sequence of dependent variables. Stochastic Processes and Appl., 83(1):79–102.
20. Parzen, E. (1963). On spectral analysis with missing observations and amplitude modulation. Sankhy¯a Ser. A, 25:383–392 (1999).
21. Rio, E. Th´eorie asymptotique des processus al´eatoires faiblement d´ependants. Collection Math´ematiques & Applications 31. Springer (2000).
22. Rosenblatt, M. A central limit theorem and a strong mixing condition. Proc. Natl. Acad. Sci. USA, 42:43–47.
23. Rosenblatt, M. (1985). Stationary Sequences and Random Fields. Birkha¨user, New York (1956).
24. Straumann, D. Estimation in conditionally heteroscedastic time series models, volume 181 of Lecture Notes in Statistics. Springer-Verlag, Berlin (2005).
25. Whittle, P. Gaussian estimation in stationary time series. Bull. Inst. Internat. Statist., 39(livraison 2):105–129 (1962).
26. Yajima, Y. and Nishino, H. Estimation of the autocorrelation function of a stationary time series with missing observations. Sankhy¯a Ser. A, 61(2):189–207 (1999).
