We investigate in this paper a Bickel–Rosenblatt test of goodness-of-fit for the density of the noise in an autoregressive model. Since the seminal work of Bickel and Rosenblatt, it is well-known that the integrated squared error of the Parzen–Rosenblatt density estimator, once correctly renormalized, is asymptotically Gaussian for independent and identically distributed (i.i.d.) sequences. We show that the result still holds when the statistic is built from the residuals of general stable and explosive autoregressive processes. In the univariate unstable case, we prove that the result holds when the unit root is located at − 1 whereas we give further results when the unit root is located at 1. In particular, we establish that except for some particular asymmetric kernels leading to a non-Gaussian limiting distribution and a slower convergence, the statistic has the same order of magnitude. We also study some common unstable cases, like the integrated seasonal process. Finally, we build a goodness-of-fit Bickel–Rosenblatt test for the true density of the noise together with its empirical properties on the basis of a simulation study.
Accepté le :
DOI : 10.1051/ps/2018016
Mots-clés : Autoregressive process, Bickel–Rosenblatt statistic, goodness-of-fit, hypothesis testing, nonparametric estimation, Parzen–Rosenblatt density estimator, residual process
@article{PS_2019__23__464_0,
author = {Lagnoux, Agn\`es and Nguyen, Thi Mong Ngoc and Pro{\"\i}a, Fr\'ed\'eric},
title = {On the {Bickel{\textendash}Rosenblatt} test of goodness-of-fit for the residuals of autoregressive processes},
journal = {ESAIM: Probability and Statistics},
pages = {464--491},
publisher = {EDP-Sciences},
volume = {23},
year = {2019},
doi = {10.1051/ps/2018016},
zbl = {1422.62284},
mrnumber = {3989599},
language = {en},
url = {http://archive.numdam.org/articles/10.1051/ps/2018016/}
}
TY - JOUR AU - Lagnoux, Agnès AU - Nguyen, Thi Mong Ngoc AU - Proïa, Frédéric TI - On the Bickel–Rosenblatt test of goodness-of-fit for the residuals of autoregressive processes JO - ESAIM: Probability and Statistics PY - 2019 SP - 464 EP - 491 VL - 23 PB - EDP-Sciences UR - http://archive.numdam.org/articles/10.1051/ps/2018016/ DO - 10.1051/ps/2018016 LA - en ID - PS_2019__23__464_0 ER -
%0 Journal Article %A Lagnoux, Agnès %A Nguyen, Thi Mong Ngoc %A Proïa, Frédéric %T On the Bickel–Rosenblatt test of goodness-of-fit for the residuals of autoregressive processes %J ESAIM: Probability and Statistics %D 2019 %P 464-491 %V 23 %I EDP-Sciences %U http://archive.numdam.org/articles/10.1051/ps/2018016/ %R 10.1051/ps/2018016 %G en %F PS_2019__23__464_0
Lagnoux, Agnès; Nguyen, Thi Mong Ngoc; Proïa, Frédéric. On the Bickel–Rosenblatt test of goodness-of-fit for the residuals of autoregressive processes. ESAIM: Probability and Statistics, Tome 23 (2019), pp. 464-491. doi : 10.1051/ps/2018016. http://archive.numdam.org/articles/10.1051/ps/2018016/
[1] and , A note on the the Bickel-Rosenblatt test in autoregressive time series. Stat. Probab. Lett. 74 (2005) 221–234. | DOI | MR | Zbl
[2] and , A sharp analysis on the asymptotic behavior of the Durbin-Watson statistic for the first-order autoregressive process. ESAIM: PS 17 (2013) 500–530. | DOI | Numdam | MR | Zbl
[3] and , On some global measures of the deviations of density function estimates. Ann. Stat. 1 (1973) 1071–1095. | DOI | MR | Zbl
[4] , Convergence of Probability Measures. Wiley Series in Probability and Statistics: Probability and Statistics. 2nd edn. John Wiley & Sons Inc., New York (1999). | DOI | MR | Zbl
[5] and , Time Series: Theory and Methods. Springer Series in Statistics. Springer, New York (2006). | MR | Zbl
[6] and , Limiting distributions of least squares estimates of unstable autoregressive processes. Ann. Stat. 16 (1988) 367–401. | MR | Zbl
[7] and . On the functional central limit theorem for stationary processes. Ann. Inst. Henri Poincaré, B. 36 (2000) 1–34. | DOI | Numdam | MR | Zbl
[8] , Random iterative models, in Vol. 34 of Applications of Mathematics. Springer-Verlag, Berlin, New York (1997). | MR | Zbl
[9] , Testing the goodness of fit of a parametric density function by kernel method. Econom. Theory 10 (1994) 316–356. | DOI | MR
[10] and , The power and optimal kernel of the Bickel-Rosenblatt test for goodness of fit. Ann. Stat. 19 (1991) 999–1009. | DOI | MR | Zbl
[11] and , Asymptotics of the Lp-norms of density estimators in the first-order autoregressive models. Stat. Probab. Lett. 66 (2004) 91–103. | DOI | MR | Zbl
[12] and , Asymptotic properties of general autoregressive models and strong consistency of least-squares estimates of their parameters. J. Multivariate Anal. 13 (1983) 1–23. | DOI | MR | Zbl
[13] and , On the Bickel-Rosenblatt test for first-order autoregressive models. Stat. Probab. Lett. 56 (2002) 23–35. | DOI | MR | Zbl
[14] and , On residual empirical processes of stochastic regression models with applications to time series. Ann. Stat. 27 (1999) 237–261. | MR | Zbl
[15] and , On bootstrapping L2-type statistics in density testing. Stat. Probab. Lett. 50 (2000) 137–147. | DOI | MR | Zbl
[16] , On estimation of a probability density function and mode. Ann. Math. Stat. 33 (1962) 1065–1076. | DOI | MR | Zbl
[17] , Further results on the H-Test of Durbin for stable autoregressive processes. J. Multivariate Anal. 118 (2013) 77–101. | DOI | MR | Zbl
[18] , Remark on some nonparametric estimates of a density function. Ann. Math. Stat. 27 (1956) 832–837. | DOI | MR | Zbl
[19] , A quadratic measure of deviation of two-dimensional density estimates and a test of independence. Ann. Stat. 3 (1975) 1–14. | DOI | MR | Zbl
[20] , Limit theorems for stochastic processes. Theory Probab. Appl. 1 (1956) 261–290. | DOI | MR | Zbl
[21] , Asymptotic properties of dynamic stochastic parameter estimates (III). J. Multivariate Anal. 4 (1974) 351–381. | DOI | MR | Zbl
[22] and , Central limit theorems for integrated square error of nonparametric density estimators based on absolutely regular random sequences. Yokohama Math. J. 35 (1987) 95–111. | MR | Zbl
[23] and , Bickel-Rosenblatt test for weakly dependent data. Math. Model. Anal. 17 (2012) 383–395. | DOI | MR | Zbl
Cité par Sources :
