zbMATH — the first resource for mathematics

Properties of the sieve bootstrap for fractionally integrated and non-invertible processes. (English) Zbl 1164.62053
This paper deals with the investigation of the consequences of applying the sieve bootstrap under regularity conditions that are sufficiently general to encompass both fractionally integrated and non-invertible processes. The sieve bootstrap is obtained by approximating the data-generating process by an autoregression, whose order $$h$$ increases with the sample size $$T$$. The author establishes the validity of the sieve bootstrap for $$| d|<1/2$$ and shows that when the sieve bootstrap is used to approximate the distribution of a general class of statistics then the error rate will be of an order smaller than $$T^{(1/2)(\beta+\max\{0,d\}-1)}$$, $$\beta>0$$. Simulation experiments examine the distribution of the maximum likelihood estimator $$\tilde d_{T}$$ of $$d$$ for the fractional noise process $$(1-z)^{d}y(t)=\varepsilon(t)$$, where $$\varepsilon(t)$$ is standard Gaussian white noise.

MSC:
 62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH) 62G09 Nonparametric statistical resampling methods
KernSmooth
Full Text:
References:
 [1] DOI: 10.1007/BF02532251 · Zbl 0202.17301 [2] DOI: 10.1017/S0266466605050383 · Zbl 1083.62080 [3] DOI: 10.1016/j.jeconom.2005.06.010 · Zbl 1345.62055 [4] DOI: 10.1111/j.1468-0262.2004.00509.x · Zbl 1141.62336 [5] Apostol T. M., Mathematical Analysis (1960) [6] DOI: 10.1016/0304-4076(95)01732-1 · Zbl 0854.62099 [7] Baxter G., Math. Scand. 10 pp 137– (1962) · Zbl 0112.09401 [8] DOI: 10.1214/ss/1177011122 [9] Beran J., Statistics for Long Memory Processes (1994) · Zbl 0869.60045 [10] DOI: 10.1214/aos/1176342709 · Zbl 0317.62064 [11] DOI: 10.1214/aos/1176345637 · Zbl 0449.62034 [12] DOI: 10.2307/3318584 · Zbl 0874.62102 [13] J. Burg (1968 ) A new analysis technique for time series data . Tech. Rep., N.A.T.O., Enschede, the Netherlands: Advanced Study Institute on Signal Processing. [14] DOI: 10.1111/1467-9868.00244 · Zbl 0966.62027 [15] DOI: 10.2307/1401322 · Zbl 0101.35604 [16] DOI: 10.2307/2335475 [17] Granger C. W. J., Journal of Time Series Analysis 1 pp 15– (1980) · Zbl 0541.62106 [18] S. Grose, and D. S. Poskitt (2006 ) The finite-sample properties of autoregressive approximations of fractionally-integrated and non-invertible processes . Tech. Rep., Department of Econometrics & Business Statistics, Working Paper 15/06, Monash University. http://www.buseco.monash.edu.au/depts/ebs/pubs/wpapers. [19] Hannan E. J., The Statistical Theory of Linear Systems (1988) · Zbl 0641.93002 [20] DOI: 10.1111/1468-0262.00439 · Zbl 1154.62361 [21] DOI: 10.1093/biomet/68.1.165 · Zbl 0464.62088 [22] DOI: 10.1016/0304-4076(95)01740-2 · Zbl 0854.62084 [23] Kolmorgorov A. N., Bulletin Academy Science U. S. S. R., Mathematics Series 5 pp 3– (1941) [24] DOI: 10.1214/aos/1176347265 · Zbl 0684.62035 [25] Levinson N., Journal of Mathematical Physics 25 pp 261– (1947) [26] DOI: 10.1017/S0266466604203024 · Zbl 1061.62022 [27] DOI: 10.1214/aos/1051027882 · Zbl 1067.62021 [28] DOI: 10.2307/1267380 · Zbl 0269.62061 [29] Munroe M. E., Introduction to Measure and Integration (1953) · Zbl 0050.05603 [30] DOI: 10.1109/TAC.1974.1100733 · Zbl 0317.62063 [31] Paulsen J., Journal of the Royal Statistical Society 47 pp 216– (1985) [32] Poskitt D. S., Annals Institute Statistical Mathematics (2006) [33] DOI: 10.1214/aos/1176344897 · Zbl 0425.62069 [34] Szego G., Orthogonal Polynomials (1939) [35] Taniguchi M., Journal of Time Series Analysis 5 pp 37– (1984) [36] DOI: 10.2307/2335553 [37] Wand M., Kernel Smoothing (1995) [38] Wold H., The Analysis of Stationary Time Series (1938) · Zbl 0019.35602 [39] DOI: 10.1111/j.1467-9892.2005.00459.x · Zbl 1115.62093
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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.