×

A new Bayesian approach to quantile autoregressive time series model estimation and forecasting. (English) Zbl 1281.62184

Summary: This paper proposes a Bayesian approach to quantile autoregressive (QAR) time series model estimation and forecasting. We establish that the joint posterior distribution of the model parameters and future values is well defined. The associated Markov chain Monte Carlo algorithm for parameter estimation and forecasting converges to the posterior distribution quickly. We also present a combining forecasts technique to produce more accurate out-of-sample forecasts by using a weighted sequence of fitted QAR models. A moving window method to check the quality of the estimated conditional quantiles is developed. We verify our methodology using simulation studies and then apply it to currency exchange rate data. The results obtained show that an unequally weighted combining method performs better than other forecasting methodology.

MSC:

62M09 Non-Markovian processes: estimation
62F15 Bayesian inference
62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
62M20 Inference from stochastic processes and prediction
62P20 Applications of statistics to economics
65C05 Monte Carlo methods
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Alston, Proceedings of the 55th International Statistical Institute (2005)
[2] Armstrong, Error measures for generalizing about forecasting methods: empirical comparisons, International Journal of Forecasting 8 pp 69– (1992) · doi:10.1016/0169-2070(92)90008-W
[3] Armstrong, Principles of Forecasting: A Handbook for Researchers and Practitioners (2001) · doi:10.1007/978-0-306-47630-3
[4] Batchlor, Forecaster diversity and the benefits of combining forecasts, Management Science 41 pp 68– (1995) · Zbl 0829.90033 · doi:10.1287/mnsc.41.1.68
[5] Bates, The combination of forecasts, Operational Research Quarterly 20 pp 451– (1969) · doi:10.1057/jors.1969.103
[6] Box, Time Series Analysis: Forecasting and Control (1994) · Zbl 0858.62072
[7] Bunn, Interaction of judgemental and statistical forecasting methods: issues and analysis, Management Science 37 pp 501– (1991) · doi:10.1287/mnsc.37.5.501
[8] Cai, Convergence theory of a numerical method for solving Chapman Kolmogorov equation, SIAM Journal on Numerical Analysis 40 pp 2337– (2003) · Zbl 1073.65013 · doi:10.1137/S0036142901390366
[9] Cai, A forecasting procedure for nonlinear autoregressive time series models, The Journal of Forecasting 24 pp 335– (2005) · doi:10.1002/for.959
[10] Cai, A quantile approach to US GNP, Economic Modelling 24 pp 969– (2007) · doi:10.1016/j.econmod.2007.03.007
[11] Cai, Quantile self-exciting autoregressive time series models, Journal of Time Series Analysis 29 pp 186– (2008) · Zbl 1165.62062 · doi:10.1111/j.1467-9892.2007.00551.x
[12] Cai, Forecasting for quantile self exciting threshold autoregressive time series models, Biometrika 97 pp 199– (2010) · Zbl 1183.62158 · doi:10.1093/biomet/asp070
[13] Clark, Combining forecasts from nested models, Oxford Bulletin of Economics and Statistics 71 pp 303– (2009) · doi:10.1111/j.1468-0084.2009.00547.x
[14] Clemen, Combining forecasts: A review and annotated bibliography, International Journal of Forecasting 5 pp 559– (1989) · doi:10.1016/0169-2070(89)90012-5
[15] Gamerman, Markov Chain Monte Carlo: Stochastic Simulation for Bayesian Inference (2002)
[16] Geyer, Handbook of Markov chain Monte Carlo pp 3– (2011)
[17] Hastings, Monte Carlo sampling method using Markov chain and their applications, Biometrika 57 pp 97– (1970) · Zbl 0219.65008 · doi:10.1093/biomet/57.1.97
[18] Koenker, Quantiles Regression (2005) · doi:10.1017/CBO9780511754098
[19] Koenker, Conditional quantile estimation and inference for ARCH models, Econometric Theory 12 pp 793– (1996) · Zbl 04535706 · doi:10.1017/S0266466600007167
[20] Koenker, Unit root quantile autoregression inference, Journal of the American Statistical Association 99 pp 775– (2004) · Zbl 1117.62370 · doi:10.1198/016214504000001114
[21] Koenker, Quantile autoregression, Journal of American Statistical Association 101 pp 980– (2006) · Zbl 1120.62326 · doi:10.1198/016214506000000672
[22] Kumar, Numerical prediction of typhoon tracks and intensity using a multimodel superensemble, Monthly Weather Review 131 pp 574– (2003)
[23] McCulloch, Bayesian inference and prediction for mean and variance shifts in autoregressive time series, Journal of the American Statistical Association 88 pp 968– (1993) · Zbl 0800.62165 · doi:10.1080/01621459.1993.10476364
[24] McCulloch, Bayesian analysis of autoregressive time series via the Gibbs sampler, Journal of Time Series Analysis 15 pp 235– (1994) · Zbl 0800.62549 · doi:10.1111/j.1467-9892.1994.tb00188.x
[25] McNees, The uses and abuses of ’consensus’ forecasts, Journal of Forecasting 11 pp 703– (1992) · Zbl 04510431 · doi:10.1002/for.3980110807
[26] Rosenberg, A Bayesian approach to understanding time series data, North American Acturial Journal 3 pp 130– (1999) · Zbl 1082.62503 · doi:10.1080/10920277.1999.10595808
[27] Tong, Non-linear Time Series, A Dynamical System Approach (1990) · Zbl 0716.62085
[28] Wallis, Combining forecasts - forty years later, Applied Financial Economics 21 pp 33– (2011) · doi:10.1080/09603107.2011.523179
[29] Wang, International Seminar on Future Information Technology and Management Engineering, 2008. FITME ’08 pp 450– (2008) · doi:10.1109/FITME.2008.42
[30] Yu, Bayesian quantile regression, Statistics and Probability Letters 54 pp 437– (2001) · Zbl 0983.62017 · doi:10.1016/S0167-7152(01)00124-9
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.