Abstract
A sequence of time-ordered observations follows an autoregressive model of order one and its parameter is possibly subject to change at most once at some unknown time point. The aim is to test whether such an unknown change has occurred or not. A change-point method presented here rely on a ratio type test statistic based on the maxima of cumulative sums. The main advantage of the developed approach is that the variance of the observations neither has to be known nor estimated. Asymptotic distribution of the test statistic under the no-change null hypothesis is derived. Moreover, we prove the consistency of the test under the alternative. A bootstrap procedure is proposed in the way of a completely data-driven technique without any tuning parameters. The results are illustrated through a simulation study, which demonstrates the computational efficiency of the procedure. A practical application to real data is presented as well.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Hušková, M., Prášková, Z., Steinebach, J.: On the detection of changes in autoregressive time series I. Asymptotics. J. Stat. Plan. Infer. 137(4), 1243–1259 (2007)
Hušková, M., Kirch, C., Prášková, Z., Steinebach, J.: On the detection of changes in autoregressive time series, II. Resampling procedures. J. Stat. Plan. Infer. 138(6), 1697–1721 (2008)
Horváth, L., Horváth, Z., Hušková, M.: Ratio tests for change point detection. In: Balakrishnan, N., Peña, E.A., Silvapulle, M.J. (eds.) Beyond Parametrics in Interdisciplinary Research: Festschrift in Honor of Professor Pranab K. Sen, vol. 1, pp. 293–304. IMS Collections, Beachwood, Ohio (2009)
Peštová, B., Pešta, M.: Testing structural changes in panel data with small fixed panel size and bootstrap. Metrika 78(6), 665–689 (2015)
Peštová, B., Pešta, M.: Change point estimation in panel data without boundary issue. Risks 5(1), 7 (2017)
Prague Stock Exchange: PX Index 2015. https://www.pse.cz/dokument.aspx?k=Burzovni-Indexy. Updated April 30, 2015; Accessed April 30, 2015
Davidson, J.: Stochastic Limit Theory: An Introduction for Econometricians. Oxford University Press, New York (1994)
Acknowledgements
Institutional support to Barbora Peštová was provided by RVO:67985807. Michal Pešta was supported by the Czech Science Foundation project No. 18-01781Y.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix: Proofs
Appendix: Proofs
Proof
(of Theorem 1) Let us consider an array
and a filtration \(\mathcal {F}_{n,i}=\sigma \{\varepsilon _j,\,j\le i\}\), \(i=2,\ldots ,n\) and \(n\in \mathbb {N}\). Then, \(\{U_{n,i},\mathcal {F}_{n,i}\}\) is a martingale difference array such that
Moreover,
Furthermore,
Since \(\{Y_{i-1}^2(\varepsilon _i^2-\sigma ^2)\}\) is a martingale difference array again with respect to \(\mathcal {F}_{n,i}\), we have under Assumption 3 from the Chebyshev’s inequality that
Similarly, as a consequence of Lemma 4.2 in [1],
Thus,
Next, for any \(\epsilon >0\),
Additionally,
for all \(t\in [0,1]\).
According to Theorem 27.14 from [7] for the martingale difference array \(\{U_{n,i}, \mathcal {F}_{n,i}\}\), where the assumptions of this theorem are satisfied due to (11), (12), and (13), we get
Therefore,
where \(\widetilde{\mathcal {W}}(t)=\mathcal {W}(1)-\mathcal {W}(t)\).
Let us define
Hence, for the expression in the numerator of \(\mathcal {V}_n\), it holds
Similarly, for the expression in the denominator of \(\mathcal {V}_n\),
Lemma 4.2 in [1] gives
and
as \(n\rightarrow \infty \), where [nt] and \([n(1-t)]\) mean truncated number to zero decimal digits. Finally, (14) together with (15), (16), (17), and (18) implies
Then, the assertion of the theorem directly follows. \(\square \)
Proof
(of Theorem 2) Let us take \(k=\tau +2\), \(k=[\xi n]\) for some \(\zeta<\xi <1-\gamma \) and \(i=\tau +1\). Then,
According to the proof of Theorem 1, as \(n\rightarrow \infty \),
Lemma 4.2 from [1] gives
Now,
For \(\tau <k=[\xi n]\), the denominator in (4) divided by \(\sqrt{n-1}\) has the same distribution as under the null hypothesis and it is, therefore, bounded in probability. It follows that the maximum of the ratio has to tend in probability to infinity as well, while \(n\rightarrow \infty \). \(\square \)
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Peštová, B., Pešta, M. (2018). Asymptotic and Bootstrap Tests for a Change in Autoregression Omitting Variability Estimation. In: Rojas, I., Pomares, H., Valenzuela, O. (eds) Time Series Analysis and Forecasting. ITISE 2017. Contributions to Statistics. Springer, Cham. https://doi.org/10.1007/978-3-319-96944-2_13
Download citation
DOI: https://doi.org/10.1007/978-3-319-96944-2_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-96943-5
Online ISBN: 978-3-319-96944-2
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)