Abstract
The time series models with normally distributed innovations generate stationary normal sequences. However, if the innovations are not normal then the stationary marginal distribution may be a member of an entirely different family. This chapter discusses autoregressive models with innovations belonging to various classes of non-Gaussian distributions. Detailed analysis of the models with innovation distributions such as stable, Laplace, heavy-tailed, exponential, gamma, and mixed normal is considered along with the problem of estimation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
See the discussions in Sect. 1.2.3.
- 3.
- 4.
See Sect. 2.7 for details on the Estimating Function methods.
- 5.
- 6.
Billingsley’s regularity conditions for the CAN properties of MLE and the related asymptotic results are stated in Sect. 2.3
- 7.
- 8.
See Sect. 2.3.3 for details on MMLE.
- 9.
An AR(1) model with stationary Levy-marginal distribution is discussed in Sect. 3.7.3.
References
B. Mandelbrot, The variation of certain speculative prices. J. Bus. 36, 394– 419 (1963)
B. Mandelbrot, The variation of some other speculative prices. J. Bus. 40, 393–413 (1967)
B. Mandelbrot, H.M. Taylor, On the distribution of stock price difference. Oper. Res. 15, 1057–1062 (1967)
E.F. Fama, The behavior of stock-market prices. J. Bus. 38(1), 34–105 (1965)
G. Schoups, J.A. Vrugt, A formal likelihood function for parameter and predictive inference of hydrologic models with correlated, heteroscedastic, and non-gaussian errors. Water Res. Res. 46, 1–17 (2010)
J. Andel, On AR (1) processes with exponential white noise. Commun. Stat. Theory Methods 17(5), 1481–1495 (1988)
C.B. Bell, E.P. Smith, Inference for non-negative autoregressive schemes. Commun. Stat. Theory Methods 15(8), 2267–2293 (1986)
P. Bondon, Estimation of autoregressive models with epsilon-skew-normal innovations. J. Multivar. Anal. 100(8), 1761–1776 (2009)
W. Hurlimann, On non-Gaussian AR (1) inflation modeling,. J. Stat. Econ. Methods 1(1), 93–101 (2012)
W.K. Li, A.I. McLeod, ARMA modelling with non-Gaussian innovations. J. Time Ser. Anal. 9(2), 155–168 (1988)
G.E.P. Box, G.M. Jenkins, G.C. Reinsel, Time Series Analysis: Forecasting and Control (Prentice Hall, Englewood Cliffs, 1994)
P.J. Brockwell, R.A. Davis, Time Series Theory and Methods, 2nd edn. (Springer, New York., 2006)
G. Samorodnitsky, M. Taqqu, Stable Non-Gaussian Random Processes (Chapman Hall, 1994)
S. Davis, R.A. Resnick, Limit theory for the sample covariance and correlation functions of moving averages. Ann. Stat. 14, 533–558 (1986)
J.W. Lin, A.I. McLeod, Portmanteau tests for ARMA models with infinite variance. J. Time Ser. Anal. 29(3), 600–617 (2008)
D.B.H. Cline, P.J. Brockwell, Linear prediction of ARMA processes with infinite variance. Stoch. Process. Appl. 19, 281–296 (1985)
Z. Qiou, N. Ravishanker, Bayesian inference for time series with stable innovations. J. Time Ser. Anal. 19(2), 235–249 (1998)
J.L. Knight, J. Yu, Empirical characteristic function in time series estimation. Economet. Theor. 18(3), 691–721 (2002)
T. Merkouris, Transform martingale estimating functions. Ann. Stat. 35(5), 1975–2000 (2007)
A. Thavaneswaran, N. Ravishanker, Y. Liang, Inference for linear and nonlinear stable error processes via estimating functions. J. Stat. Plann. Inference. 143(4), 827–841 (2013)
B. Andrews, M. Calder, R.A. Davis, Maximum likelihood estimation for \(\alpha -\)stable autoregressive processes. Ann. Stat. 37(4), 1946–1982 (2009)
R.M. Zhang, N.H. Chan, Maximum likelihood estimation for nearly non-stationary stable autoregressive processes. J. Time Ser. Anal. 33, 542–553 (2012)
P. Billingsley, Statistical inference for Markov processes, vol. 2 (University of Chicago Press, 1961)
M. Shao, C.L. Nikias, Signal processing with fractional lower ordermoments: stable processes and their applications. Proc IEEE. 81(7), 986–1010 (1993)
C.M. Gallagher, A method for fitting stable autoregressive models using the autocovariation function. Stat. Probab. Lett. 53, 381–390 (2001)
N. Balakrishna, G. Hareesh, Analysis of autoregressive models with symmetric stable innovations. Statistics 52(2), 288–302 (2018)
N. Balakrishna, G. Hareesh, Statistical signal extraction using stable processes. Stat. Probab. Lett. 79(7), 851–856 (2009)
N. Balakrishna, G. Hareesh, Stable autoregressive models and signal estimation. Commun. Stat. Theory Methods 41(11), 1969–1988 (2012)
J.P. Nolan, N. Ravishanker, Simultaneous prediction intervals for ARMA processes with stable innovations. J. Forecast. 28, 235–246 (2009)
N.L. Johnson, S. Kotz, N. Balakrishnan, Continuous Univariate Distributions (Wiley, New York., 1995)
N. Balakrishna, C.K. Nampoothiri, Cauchy autoregressive process and its applications. J. Indian Stat. Assoc. 41(2), 143–156 (2003)
J. Andel, T. Barton, A note on threshold AR(1) model with cauchy innovations. J. Time Ser. Anal. 7, 1–5 (1986)
J. Choi, I. Choi, Maximum likelihood estimation of autoregressive models with a near unit root and cauchy errors. Ann. Inst. Stat. Math. 71, 1121–1142 (2019)
E. Damsleth, A.H. El-Shaarawi, ARMA models with double-exponentially distributed noise. J. R. Stat. Soc. B 51(1), 61–69 (1989)
S.Y. Hwang, I.V. Basawa, Godambe estimating functions and asymptotic optimal inference. Stat. Probab. Lett. 81(8), 1121–1127 (2011)
P.H. Diananda, Note on some properties of maximum likelihood estimates. Proc. Camb. Philos. Soc. 45, 536–544 (1949)
M.K. Varanasi, B. Aazhang, Parametric generalized gaussian density estimation. J. Acoust. Soc. Am. 86(4), 1404–1415 (1989)
S. Nadarajah, A generalized normal distribution. J. Appl. Stat. 32(7), 685–694 (2005)
G. Agro, Maximum likelihood estimation for the exponential power function parameters. Commun. Stat. Simul. Comput. 24, 523–536 (1995)
G.E.P. Box, G.C. Tiao, A further look at robustness via Bayes’s theorem. Biometrika 49, 419–432 (1962)
J. Ledolter, Inference robustness of ARIMA models under non-normality - special application to stock price data. Metrika 26, 43–56 (1979)
R.W. Barnard, A.A. Trindade, R.I.P. Wickamasinghe, Autoregressive moving average models under exponential power distributions. Prob. Stat. Forum 07, 65–77 (2014)
N. Balakrishna, C.G. Sri Ranganath, ARMA models with generalized error distributed innovations. J. Indian Stat. Assoc. 53(1), 11–34 (2015)
M.L. Tiku, W.K. Wong, D.C. Vaughan, G. Bian, Time series models in non-normal situations: symmetric innovations. J. Time Ser. Anal. 21(5), 571–596 (2000)
H. Haghbin, A.R. Nematollahi, Likelihood-based inference in autoregressive models with scaled t-distributed innovations by means of em-based algorithms. Commun. Stat. - Simul. Comput. 42(10), 2239–2252 (2013)
U.C. Nduka, Em-based algorithms for autoregressive models with t-distributed innovations. Commun. Stat. Simul. Comput. 47(1), 206–228 (2018)
M.L. Tiku, R.P. Suresh, A new method of estimation for location and scale parameters. J. Stat. Plann. Inference 30, 281–292 (1992)
W. Polasec, J. Pai, Cpo plots for ARMA model selection. CEJOR 12, 211–219 (2004)
A.D. Akkaya, M.L. Tiku, Time series AR(1) model for shorttailed distributions. Stat: A J. Theor. Appl. Stat. 39(2), 117 – 132 (2005)
M.L. Tiku, W.K. Wong, G. Bian, Time series models with asymmetric innovations. Commun. Stat. Theory and Methods 28(6), 1331–1360 (1999)
W.K. Wong, G. Bian, Estimating parameters in autoregressive models with asymmetric innovations. Stat. Probab. Lett. 71, 61–70 (2005)
T.A. Ula, C. Yozgatligil, Modified maximum-likelihood method for non-normal time series revisited. Commun. Stat. Theory Method. 33(2), 397–417 (2004)
N. Sarlac, Annual streamflow modelling with asymmetric distribution function. Hydrol. Process. 22, 3403–3409 (2008)
A. Trindade, Y. Zhu, B. Andrews, Time series models with asymmetric Laplace innovations. J. Stat. Comput. Simul. 80(12), 1317–1333 (2010)
D.S. Mudholkar, A.D. Hutson, The epsilon-skew-normal distribution for analyzing near-normal data. J. Stat. Plann. Inference 83(2), 291–309 (2000)
O. Barndorff-Nielsen, Exponentially decreasing distributions for the logarithm of particle size, in Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences, vol. 353 (The Royal Society, 1977), pp. 401–419
O. Barndorff-Nielsen, N. Shephard, Non-Gaussian ornstein-uhlenbeck-based models and some of their uses in financial economics (with discussion), J. R. Statist. Soc. B 63, 167–241. The Royal Society (2001)
C. Halgreen, Self-decomposability of the generalized inverse Gaussian and hyperbolic distributions. Z. Wahrsch. verw. Gebiete 71(1), 13–17 (1979)
S. Ghasami, Z. Khodadadi, M. Maleki. Autoregressive processes with generalized hyperbolic innovations. Commun. Stat. Simul. Comput. 1–14 (2019)
H. Karttunen, An autoregressive model based on the generalized hyperbolic distribution. Scand. J. Stat. 47, 787–816 (2020)
O. Barndorff-Nielsen, Normal inverse Gaussian distributions and stochastic volatility modelling. Scand. J. Stat. 24(1), 1–13 (1997)
G.N. Boshnakov, Analytic expressions for predictive distributions in mixture autoregressive models. Statist. Probab. Lett. 79(15), 1704–1709 (2009)
N.D. Le, R.D. Martin, A.E. Raftery, Modeling flat stretches, bursts, and outliers in time series using mixture transition distribution models. J. Amer. Statist. Assoc. 91, 1504–1515 (1996)
C.S. Wong, W.K. Li, On a mixture autoregressive model. J. Roy. Stat. Soc. B 62(1), 91–115 (2000)
C.S. Wong, W.S. Chan, Mixture Gaussian time series modeling of long-term market returns. North American Actuarial Journal. 9(4), 83–94 (2013)
C. S. Wong, W. S. Chan, and Kam P. L. A student t-mixture autoregressive model with applications to heavy-tailed financial data. Biometrika., 96:751–760, (2009)
H.D. Nguyena, G.J. McLachlan, J.F.P. Ullmannb, A.L. Janke, Laplace mixture autoregressive models. Stat. Probab. Lett. 110, 18–24 (2016)
S. **, W.K. Li, Modeling panel time series with mixture autoregressive model. J. Data Sci. 4, 425–446 (2006)
A. Aknouche, Recursive online em estimation of mixture autoregressions. J. Stat. Comput. Simul. 83(2), 370–383 (2013)
L. Kalliovirta, M. Meitz, P. Saikkonen, A Gaussian mixture autoregressive model. J. Time Series Anal. 36, 247–266 (2015)
M. Meitz, P. Saikkonen, Testing for observation-dependent regime switching in mixture autoregressive model. J. Econ. 222(1), 601–624 (2021)
J.D. Hamilton, A new approach to the economic analysis of nonstationary time series and the business cycle. Econometrica 57(2), 357–384 (1989)
R.L.G. Latimier, E.L. Bouedec, V. Monbet, Markov switching autoregressive modeling of wind power forecast errors. Electr. Power Syst. Res. 189, 1–7 (2020)
B. Nielsen, N. Shephard, Likelihood analysis of a first-order autoregressive model with exponential innovations. J. Time Ser. Anal. 24, 337–344 (2003)
H. Fellag, L. Atil, C. Belkacem, On the stability of estimation of AR(1) coefficient in the presence of contaminated exponential innovations. Afr. Stat. 9, 647–658 (2014)
R.A. Davis, W.P. McCormick, Estimation for first- order autoregressive processes with positive or bounded innovations. Stoch. Process. Appl. 31, 237–250 (1989)
P.D. Feigin, S.I. Resnick, Estimation for autoregressive processes with positive innoatins. Commun. Stat. Stoch. Models 8, 479–498 (1999)
B. Abraham, N. Balakrishna, Inverse Gaussian autoregressive models. J. Time Ser. Anal. 20(6), 605–618 (1999)
Ing Chiang-Kang, Yang Chiao-Yi, Predictor selection for positive autoregressive processes. J. Am. Stat. Assoc. 109(505), 243–253 (2014)
S. Datta, W.P. McCormick, Bootstrap inference for a first-order autoregression with positive innovations. J. Am. Stat. Assoc. 90(432), 1289–1300 (1995)
Hsian Wei-Cheng, Huang Hao-Yun, Ing Chiang-Kang, Interval estimation for a first order positive autoregressive process. J. Time Ser. Anal. 39, 447–467 (2018)
A. Bartlett, W.P. McCormick, Estimation for non-negative time series with heavy-tail innovations. J. Time Ser. Anal. 34, 96–115 (2013)
P.D. Feigin, S.I. Resnick, C. Starica, Testing for independence in heavy tailed and positive innovation time series. Commun. Stat. Stoch. Models 11(7), 587–612 (1995)
J. Andel, Non-negative autoreressive processes. J. Time Ser. Anal. 10(1), 1–11 (1989)
K.H. Lee, J.G. Park, Estimation in autoregressive process with non-negative innovations. Youngnam Stat. Lett. 3(1), 65–78 (1992)
P.D. Feigin, S.I. Resnick, Limit distributions for linear programming time series estimators. Stoch. Process. Their Appl. 51, 135–165 (1994)
B. Lagos-Álvarez, G. Ferreira1, E. Porcu, Modified maximum likelihood estimation in autoregressive processes with generalized exponential innovations. Open J. Stat. 4, 620–629 (2014)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Balakrishna, N. (2021). Linear Time Series Models with Non-Gaussian Innovations. In: Non-Gaussian Autoregressive-Type Time Series. Springer, Singapore. https://doi.org/10.1007/978-981-16-8162-2_6
Download citation
DOI: https://doi.org/10.1007/978-981-16-8162-2_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-8161-5
Online ISBN: 978-981-16-8162-2
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)