
Nonstationary Time Series Prediction and Structural Break Points Diagnosis for Inflation: A Case Study of China, USA and Other Countries
GAO Weiqing, WU Ben, ZHANG Bo
China Journal of Econometrics ›› 2023, Vol. 3 ›› Issue (1) : 108-127.
Nonstationary Time Series Prediction and Structural Break Points Diagnosis for Inflation: A Case Study of China, USA and Other Countries
China and the world today are undergoing great changes that have not been seen in a century, and it is fundamental to maintain the country's economic stability. As is known, extreme inflation is one of the major sources of economic instability. Modeling and predicting inflation thus become an urgent problem to be solved. In this work, we investigate the consumer price index (CPI) inflation from the recent ten years of four major countries in the world, including China and USA, and propose a novel stochastic volatility in mean model with time-varying parameters and structural breaks in the volatility (SB-TVP-SVM), and a corresponding Bayesian inference framework. In most of the previous studies, researchers usually ignored the potential coexistent non-stationaritiy of both the conditional mean and volatility series for CPI inflation. By introducing unobserved structural break points in the volatility process, SB-TVP-SVM model solved this problem and achieves better prediction accuracy than its competitors. The structural break points estimated from our model are found to be highly related to the biggest global events over the last decade such as the COVID-19 epidemic and the conflict between Russia and Ukraine.
inflation / piecewise stable / structural break points / Bayesian estimation / TVP-SVM {{custom_keyword}} /
表1 中国、美国、德国和韩国四国2012年2月至2022年7月的月度CPI通胀率的基本统计特征 |
国家 | 均值 | 标准差 | 最小值 | 最大值 | 偏度 | 峰度 |
中国 | 0.00 | 7.42 | 19.23 | 1.37 | ||
美国 | 0.82 | 1.22 | 5.25 | 0.54 | 2.58 | |
德国 | 1.04 | 2.93 | 12.92 | 0.45 | 3.88 | |
韩国 | 0.95 | 6.04 | 20.42 | 0.32 | 0.71 |
图2 基于SB-TVP-SVM检验中国、美国、德国和韩国2012年2月至2022年7月的月度CPI通胀率的结构性断点. 图(a)是 |
表2 SB-TVP-SVM、LBI、CUSUM、Student、Cramer-von-Mises五种方法下中国、美国、德国、韩国CPI通胀率的结构性断点诊断 |
国家 | 方法 | 日期 |
中国 | SB-TVP-SVM | 2/2017, 10/2019, 7/2022 |
LBI | 3/2019, 5/2020 | |
CUSUM | 2/2017, 6/2020 | |
Student | 4/2012, 6/2012, 1/2014, 3/2015, 1/2016, 1/2017, 1/2018, 1/2019, 2/2020 | |
Cramer-von-Mises | 8/2012, 1/2014, 2/2105, 1/2016, 1/2017, 1/2018, 1/2019, 9/2019, 6/2020, 4/2021 | |
美国 | SB-TVP-SVM | 11/2013, 3/2020, 6/2022 |
LBI | 10/2014, 8/2020 | |
CUSUM | 1/2015, 6/2020 | |
Student | 5/2012, 10/2013, 1/2015, 7/2015, 7/2018, 11/2018, 3/2020, 10/2020 | |
Cramer-von-Mises | 8/2012, 8/2013, 1/2015, 7/2015, 8/2016, 3/2020, 10/2020 | |
德国 | SB-TVP-SVM | 10/2012, 9/2019, 6/2020, 4/2022 |
LBI | 5/2019, 9/2020 | |
CUSUM | 12/2014, 7/2020 | |
Student | 10/2014, 11/2015, 1/2017, 12/2018, 6/2020, 12/2020, 3/2022 | |
Cramer-von-Mises | 6/2014, 11/2014, 11/2015, 10/2016, 2/2017, 4/2018, 2/2019, 1/2020, 12/2020, 12/2021 | |
韩国 | SB-TVP-SVM | 9/2012 |
LBI | / | |
CUSUM | / | |
Student | 8/2012, 8/2014, 7/2015, 8/2016, 9/2017, 7/2018, 9/2019, 7/2020, 10/2021 | |
Cramer-von-Mises | 9/2012, 3/2013, 7/2014, 5/2015, 4/2016, 3/2017, 1/2018, 7/2018, 7/2019, 5/2020, 10/2021 |
表3 SB-TVP-SVM、TVP-SVM模型下中国、美国、德国、韩国CPI通胀率样本内估计的MAE、RMSE和LL |
方法 | MAE | RMSE | LL | MAE | RMSE | LL | |
中国 | 美国 | ||||||
SB-TVP-SVM | 2.422 | 2.956 | 0.104 | 0.129 | |||
TVP-SVM | 2.435 | 3.128 | 0.604 | 0.803 | |||
德国 | 韩国 | ||||||
SB-TVP-SVM | 1.199 | 1.576 | 2.259 | 2.754 | |||
TVP-SVM | 1.259 | 1.673 | 2.280 | 2.851 |
表4 SB-TVP-SVM、TVP-SVM、MA模型下中国、美国、德国、韩国CPI通胀率样本外预测的MAE、RMSE |
方法 | MAE | RMSE | MAE | RMSE | MAE | RMSE | MAE | RMSE | |||
中国 | 美国 | 德国 | 韩国 | ||||||||
SB-TVP-SVM | 6.799 | 8.377 | 1.398 | 1.664 | 2.853 | 3.833 | 2.882 | 3.627 | |||
TVP-SVM | 7.029 | 8.630 | 1.864 | 2.232 | 3.446 | 4.689 | 2.898 | 3.647 | |||
MA | 7.388 | 9.504 | 1.609 | 1.981 | 3.131 | 4.332 | 3.207 | 3.870 |
表5 SB-TVP-SVM、TVP-SVM和MA模型下中国、美国、德国、韩国CPI通胀率多期样本外预测的MAE和RMSE |
方法 | MAE | RMSE | MAE | RMSE | MAE | RMSE | MAE | RMSE | |||
中国 | 美国 | 德国 | 韩国 | ||||||||
SB-TVP-SVM | 6.838 | 7.756 | 1.410 | 1.584 | 3.002 | 3.588 | 2.936 | 3.353 | |||
TVP-SVM | 6.750 | 7.671 | 1.851 | 2.062 | 3.342 | 4.217 | 2.927 | 3.296 | |||
MA | 7.048 | 8.339 | 1.611 | 1.815 | 3.148 | 3.902 | 3.195 | 3.546 | |||
DM(TVP-SVM) | 3.886*** | 2.393** | 1.292 | 1.468 | |||||||
DM(MA) | 0.837 | 1.278 | 2.830*** | 2.359** | 1.111 | 1.385 | 0.800 | ||||
SB-TVP-SVM | 7.064 | 7.692 | 1.378 | 1.500 | 2.815 | 3.238 | 2.916 | 3.160 | |||
TVP-SVM | 6.829 | 7.412 | 1.844 | 1.970 | 3.256 | 3.945 | 2.896 | 3.115 | |||
MA | 7.172 | 7.834 | 1.619 | 1.737 | 3.137 | 3.625 | 3.102 | 3.348 | |||
DM(TVP-SVM) | 4.150*** | 3.740*** | 1.920* | 1.842* | |||||||
DM(MA) | 0.347 | 0.437 | 2.719*** | 2.242** | 1.835* | 1.779* | 0.912 | 0.633 |
何启志, 姚梦雨, 中国通胀预期测度及时变系数的菲利普斯曲线[J]. 管理世界, 2017, (5): 66- 78.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
郇松桦, 刘秀丽, 一种改进的CEEMDAN-LSTM-Neural Prophet Net模型: 用于COVID-19背景下我国月度消费预测[J]. 计量经济学报, 2022, 2 (3): 620- 643.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
姜婷凤, 汤珂, 刘涛雄, 基于在线大数据的通货膨胀"现时"预测[J]. 计量经济学报, 2022, 2 (3): 597- 619.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
雷立坤, 余江, 魏宇, 赖晓东, 经济政策不确定性与我国股市波动率预测研究[J]. 管理科学学报, 2018, 21 (6): 88- 98.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
刘小军, 汪寿阳, 谢海滨, 已实现波动率预测: 非对称二次滑动平均模型[J]. 计量经济学报, 2022, 2 (4): 930- 945.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
孙克娟, 肖皓, 毕慧敏, 赖明勇, 基于GVC-CGE模型的投入产出表模拟更新与全球价值链重构测度: 以RCEP为例[J]. 计量经济学报, 2022, 2 (4): 773- 795.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
习近平, (2022). 高举中国特色社会主义伟大旗帜为全面建设社会主义现代化国家而团结奋斗--- 在中国共产党第二十次全国代表大会上的报告[R]. 新华网.
Xi J P, (2022). Hold High the Great Banner of Socialism with Chinese Characteristics and Strive in Unity to Build a Modern Socialist Country in All Respects -- Report on the 20th National Congress of the CPC[R]. Xinhua.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
中国人民银行, (2022). 易纲行长在博鳌亚洲论坛的讲话[R].
People's Bank of China, (2022). PBC Governor Yi Gang's speech at the Boao Forum for Asia[R].
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
IMF, (2022). World Economic Outlook[R].
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
/
〈 |
|
〉 |