|
|
Ljung-Box检验及其改进方法的比较研究
|
Abstract:
在时间序列模型的建模过程中，需要对模型拟合的充分性进行检验，通常会检验是否存在残差自相关。目前，Ljung-Box检验是被广泛应用于各种时间序列模型建模分析中，检验残差序列是否存在自相关的方法。该方法被证实存在一些缺陷，后继学者们提出了一些新的改进方法。选取了四种基于Ljung-Box检验改进方法，通过蒙特卡洛模拟对其有限样本情况下的性质和表现进行了研究，指出了这些方法在实际应用中需要注意的问题。
In the modeling process of time series model, it is necessary to test the adequacy of model fitting, usually to test whether there is residual autocorrelation. At present, Ljung-box test is widely used in modeling and analysis of various time series models to test whether there is autocorrelation in re-sidual series. This method was proved to have some defects, and some new improvement methods were put forward by subsequent scholars. Four improved methods based on Ljung-box test are se-lected, and their properties and performance in the case of finite samples are studied by Monte Carlo simulation, and the problems needing attention in practical application of these methods are pointed out.
| [1] | Pierce, B.D.A. (1970) Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Se-ries Models. Journal of the American Statal Association, 72, 397-402. |
| [2] | Ljung, G.M. and Box, G.E.P. (1978) On a Measure of Lack of Fit in Time Series Models. Biometrika, 65, 297-303.
https://doi.org/10.1093/biomet/65.2.297 |
| [3] | Mcleod, A.I. (1978) On the Distribution of Residual Autocorrelations in Box-Jenkins Models. Journal of the Royal Statistical Society, 40, 296-302. https://doi.org/10.1111/j.2517-6161.1978.tb01042.x |
| [4] | Lee, T. (2016) Wild Bootstrap Ljung-Box Test for Cross Correlations of Multivariate Time Series. Economics Letters, 147, 59-62. https://doi.org/10.1016/j.econlet.2016.08.015 |
| [5] | Ljung, G.M. (1986) Diagnostic Testing of Univariate Time Se-ries Models. Biometrika, 73, 725-730.
https://doi.org/10.1093/biomet/73.3.725 |
| [6] | Battaglia, F. (1990) Approximate Power of Portmanteau Tests for Time Series. Statistics & Probability Letters, 9, 337-341. https://doi.org/10.1016/0167-7152(90)90143-U |
| [7] | Thoughts on the Ljung-Box Test. https://robjhyndman.com/hyndsight/ljung-box-test |
| [8] | Hassani, H. and Yeganegi, M.R. (2019) Sum of Squared ACF and the Ljung-Box Statistics. Physica A: Statistical Mechanics and Its Applications, 520, 81-86. https://doi.org/10.1016/j.physa.2018.12.028 |
| [9] | Hassani, H. and Yeganegi, M.R. (2020) Selecting Optimal Lag Order in Ljung-Box Test. Physica A: Statistical Mechanics and Its Applications, 541, 123700. https://doi.org/10.1016/j.physa.2019.123700 |
| [10] | Monti, A.C. (1994) A Proposal for a Residual Autocorrelation Test in Linear Models. Biometrika, 81, 776-780.
https://doi.org/10.1093/biomet/81.4.776 |
| [11] | Mahdi, E. and Mcleod, A.I. (2012) Improved Multivariate Portman-teau Test. Journal of Time Series Analysis, 33, 211-222. https://doi.org/10.1111/j.1467-9892.2011.00752.x |
| [12] | Pena, D. and Rodriguez, J. (2002) A Powerful Portmanteau Test of Lack of Fit for Time Series. Journal of the American Statistical Association, 97, 601-611. https://doi.org/10.1198/016214502760047122 |
| [13] | Pe？a, D. and Rodríguez, J. (2006) The Log of the Determinant of the Autocorrelation Matrix for Testing Goodness of Fit in Time Series. Journal of Statistical Planning and Inference, 136, 2706-2718.
https://doi.org/10.1016/j.jspi.2004.10.026 |
| [14] | Lin, J.W. and Mcleod, A.I. (2006) Improved Pena-Rodriguez Port-manteau Test. Computational Statistics & Data Analysis, 51, 1731-1738. https://doi.org/10.1016/j.csda.2006.06.010 |
| [15] | Box, E.P.G. (1954) Some Theorems on Quadratic Forms Applied in the Study of Analysis of Variance Problems, I. Effect of Inequality of Variance in the One-Way Classification. Annals of Mathematical Statistics, 25, 290-302.
https://doi.org/10.1214/aoms/1177728786 |
| [16] | Fisher, T.J. and Gallagher, C.M. (2012) New Weighted Portman-teau Statistics for Time Series Goodness of Fit Testing. JASA: Journal of the American Statistical Association, 107, 777-787. https://doi.org/10.1080/01621459.2012.688465 |
| [17] | Davidson, R. and Mackinnon, J.G. (1999) The Size Distortion of Bootstrap Tests. Econometric Theory, 15, 361-376.
https://doi.org/10.1017/S0266466699153040 |
