×

Nonlinearity testing and modeling for threshold moving average models. (English) Zbl 1514.62693

Summary: In this paper, we suggest a simple test and an easily applicable modeling procedure for threshold moving average (TMA) models. Firstly, based on the fitted residuals by maximum likelihood estimate (MLE) for MA models, we construct a simple statistic, which is obtained by linear arrange regression and follows \(F\)-distribution approximately, to test for threshold nonlinearity and specify the threshold variables. And then, we use some scatterplots to identify the number and locations of the potential thresholds. Finally, with the statistic and Akaike information criterion, we propose the procedure to build TMA models. Both the power of test statistic and the convenience of modeling procedure can work very well demonstrated by simulation experiments and the application to a real example.

MSC:

62-XX Statistics

Software:

FinTS
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] H. Akaike, A new look at the statistical model identification, IEEE Trans. Autom. Control 19 (1974), pp. 716-723. doi: 10.1109/TAC.1974.1100705 · Zbl 0314.62039 · doi:10.1109/TAC.1974.1100705
[2] R. Baragona, F. Battaglia, and D. Cucina, Estimating threshold subset autoregressive moving-average models by genetic algorithms, Metron LXII(1) (2004), pp. 39-61. · Zbl 1416.62476
[3] R. Baragona, F. Battaglia, and D. Cucina, Fitting piecewise linear threshold autoregressive models by means of genetic algorithms, Comput. Stat. Data Anal. 47 (2004), pp. 277-295. doi: 10.1016/j.csda.2003.11.003 · Zbl 1429.62383 · doi:10.1016/j.csda.2003.11.003
[4] P. Billingsley, The Lindeberg-Levy theorem for martingales, Proc. Amer. Math. Soc. 12 (1961), pp. 788-792. · Zbl 0129.10701
[5] P.J. Brockwell, J. Liu, and R.L. Tweedie, On the existence of stationary threshold autoregressive moving-average processes, J. Time Ser. Anal. 13 (1992), pp. 95-107. doi: 10.1111/j.1467-9892.1992.tb00096.x · Zbl 0755.62064 · doi:10.1111/j.1467-9892.1992.tb00096.x
[6] K.S. Chan, Testing for threshold autoregression, Ann. Statist. 18 (1990), pp. 1886-1894. doi: 10.1214/aos/1176347886 · Zbl 0711.62074 · doi:10.1214/aos/1176347886
[7] K.S. Chan, J.D. Petruccelli, H. Tong, and S.W. Woolford, A multiple-threshold AR(1) model, J. Appl. Probab 22 (1985), pp. 267-279. doi: 10.2307/3213771 · Zbl 0579.62074 · doi:10.2307/3213771
[8] K.S. Chan and H. Tong, On estimating thresholds in autoregressive models, J. Time Ser. Anal. 7 (1986), pp. 179-190. doi: 10.1111/j.1467-9892.1986.tb00501.x · Zbl 0596.62085 · doi:10.1111/j.1467-9892.1986.tb00501.x
[9] K.-S. Chan and H. Tong, A note on the invertibility of nonlinear ARMA models, J. Statist. Plann. Inference 140 (2010), pp. 3709-3714. doi: 10.1016/j.jspi.2010.04.036 · Zbl 1233.62154 · doi:10.1016/j.jspi.2010.04.036
[10] J.G. De Gooijer, On threshold moving-average models, J. Time Ser. Anal. 19 (1998), pp. 1-18. doi: 10.1111/1467-9892.00074 · Zbl 0902.62103 · doi:10.1111/1467-9892.00074
[11] J.E. Ertel and E.B. Fowlkes, Some algorithms for linear spline and piecewise multiple linear regression, J. Amer. Statist. Assoc. 71 (1976), pp. 640-648. doi: 10.1080/01621459.1976.10481540 · Zbl 0343.62058
[12] G.C. Goodwin and R.L. Payne, Dynamic System Identification: Experiment Design and Data Analysis, Academic Press, New York, 1977. · Zbl 0578.93060
[13] V. Haggan, S.M. Heravi, and M.B. Priestley, A study of the application of state-dependent models in non-linear time series analysis, J. Time Ser. Anal. 5 (1984), pp. 69-102. doi: 10.1111/j.1467-9892.1984.tb00379.x · Zbl 0555.62071 · doi:10.1111/j.1467-9892.1984.tb00379.x
[14] E.J. Hannan and J. Rissanen, Recursive estimation of mixed autoregressive-moving average order, Biometrika 69(1) (1982), pp. 81-94. Correction: 70(1), 303. doi: 10.1093/biomet/69.1.81 · Zbl 0494.62083 · doi:10.1093/biomet/69.1.81
[15] B.E. Hansen, Sample Splitting and Threshold Estimation, Econometrica 68 (2000), pp. 575-603. doi: 10.1111/1468-0262.00124 · Zbl 1056.62528 · doi:10.1111/1468-0262.00124
[16] D.M. Keenan, A Tukey nonadditivity-type test for time series nonlinearity, Biometrika 72 (1985), pp. 39-44. doi: 10.1093/biomet/72.1.39 · Zbl 0562.62077 · doi:10.1093/biomet/72.1.39
[17] T.L. Lai and C.Z. Wei, Least squares estimates in stochastic regression models with applications to identification and control of dynamic systems, Ann. Statist. 10 (1982), pp. 154-166. doi: 10.1214/aos/1176345697 · Zbl 0488.62071 · doi:10.1214/aos/1176345697
[18] G. Li, A note on moving-average models with feedback, J. Time Ser. Anal. 33 (2012), pp. 873-879. doi: 10.1111/j.1467-9892.2012.00802.x · Zbl 1282.62200 · doi:10.1111/j.1467-9892.2012.00802.x
[19] G. Li and W.K. Li, Testing for the threshold in moving average with conditional heteroscedasticity, Statist. Sin. 18 (2008), pp. 647-665. · Zbl 1135.62071
[20] D. Li, S. Ling, and H. Tong, On moving-average models with feedback, Bernoulli 18 (2012), pp. 735-745. doi: 10.3150/11-BEJ352 · Zbl 1456.62204 · doi:10.3150/11-BEJ352
[21] S. Ling, On the probabilistic properties of a double threshold ARMA conditional heteroskedastic model, J. Appl. Probab. 36 (1999), pp. 688-705. doi: 10.1239/jap/1032374627 · Zbl 1109.62340 · doi:10.1239/jap/1032374627
[22] S. Ling and H. Tong, Testing a linear moving-average model against threshold moving-average models, Ann. Statist. 33 (2005), pp. 2529-2552. doi: 10.1214/009053605000000598 · Zbl 1085.62102 · doi:10.1214/009053605000000598
[23] S. Ling, H. Tong, and D. Li, Ergodicity and invertibility of threshold moving-average models, Bernoulli 13 (2007), pp. 161-168. doi: 10.3150/07-BEJ5147 · Zbl 1111.62079 · doi:10.3150/07-BEJ5147
[24] J. Liu and E. Susko, On strict stationarity and ergodicity of a non-linear ARMA model, J. Appl. Probab. 29 (1992), pp. 363-373. doi: 10.2307/3214573 · Zbl 0753.62059 · doi:10.2307/3214573
[25] M.B. Priestley, State-dependent models: a general approach to non-linear time series analysis, J. Time Ser. Anal. 1 (1980), pp. 47-71. doi: 10.1111/j.1467-9892.1980.tb00300.x · Zbl 0496.62076 · doi:10.1111/j.1467-9892.1980.tb00300.x
[26] H. Tong, On a threshold model, in Pattern Recognition and Signal Processing, C.H. Chen, (eds.), Sijthoff and Noordhoff, Amsterdam, pp. 101-141, 1978.
[27] H. Tong, Non-Linear Time Series: A Dynamical System Approach, Oxford University Press, Oxford, 1990. · Zbl 0716.62085
[28] H. Tong and K.S. Lim, Threshold autoregressions, limit cycles, and data, J. R. Statist. Soc. B 42 (1980), pp. 245-292. · Zbl 0473.62081
[29] R.S. Tsay, Nonlinearity tests for time series, Biometrika 73 (1986), pp. 461-466. doi: 10.1093/biomet/73.2.461 · Zbl 0603.62097 · doi:10.1093/biomet/73.2.461
[30] R.S. Tsay, Conditional heteroscedastic time series models, J. Amer. Statist. Assoc. 82 (1987), pp. 590-604. doi: 10.1080/01621459.1987.10478471 · Zbl 0636.62092
[31] R.S. Tsay, Testing and modeling threshold autoregressive processes, J. Amer. Statist. Assoc. 84 (1989), pp. 231-240. doi: 10.1080/01621459.1989.10478760 · Zbl 0683.62050
[32] R.S. Tsay, Analysis of Financial Time Series, 2nd ed., John Wiley & Sons, New York, 2005. · Zbl 1086.91054 · doi:10.1002/0471746193
[33] C.S. Wong and W.K. Li, Testing for threshold autoregression with conditional heteroscedasticity, Biometrika 84 (1997), pp. 407-418. doi: 10.1093/biomet/84.2.407 · Zbl 1058.62554 · doi:10.1093/biomet/84.2.407
[34] C.S. Wong and W.K. Li, Testing for double threshold autoregressive conditional heteroscedastic model, Statist. Sin. 10 (2000), pp. 173-189. · Zbl 0970.62059
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. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.