基于卷积长短时记忆网络的CPI预测

doi:10.3778/j.issn.1002-8331.2011-0227

摘要/Abstract

摘要： 针对消费价格指数（CPI）的预测值滞后于真实值的现象，提出一种基于卷积神经网络-长短期记忆（CNN-LSTM）深度网络的CPI预测模型，预测结果相较于传统方法有较小的均方根误差和平均绝对百分比误差，且预测结果的定向精度和Pearson相关系数显著高于传统方法。用卷积神经网络-长短期记忆深度网络学习期货数据的空间特征和时间特征，动态定量预测每日CPI的变化情况。为有效提高深度网络训练的样本数量，对月度CPI数据进行数据增强。通过滑动时间窗口动态训练模型，预测2019年1月至2020年5月CPI变化情况。模型预测CPI取得了较高的准确率，在基于日级别数据进行CPI预测时具有明显优势。

关键词: CPI预测, CNN-LSTM深度网络, 面板数据, 数据增强, 动态预测

Abstract: Aiming at the phenomenon that the predicted value of consumer price index（CPI） lags behind the real value, a CPI prediction model based on the convolutional neural network-long-and short-term memory（CNN-LSTM） deep network is proposed. Compared with the traditional method, the prediction result has smaller root mean square error and average absolute percentage error, and the orientation accuracy and Pearson correlation coefficient of the prediction result are significantly higher than those of the traditional method. Firstly, the convolution neural network-long-term and short-term memory depth network is used to learn the spatial and temporal characteristics of futures data, and dynamically and quantitatively predict the changes of daily CPI. Then, in order to effectively improve the number of samples of in-depth network training, the monthly CPI data are enhanced. Finally, through the dynamic training model of sliding time window, the change of CPI from January 2019 to May 2020 is predicted. The model achieves high accuracy in CPI prediction and has obvious advantages in CPI prediction based on daily level data.

Key words: CPI forecasting, convolutional neural network-long-and short-term memory（CNN-LSTM） deep neural network, panel data, data augmentation, dynamic forecasting

陈逸东, 陆忠华. 基于卷积长短时记忆网络的CPI预测[J]. 计算机工程与应用, 2022, 58(9): 256-262.

CHEN Yidong, LU Zhonghua. Forecasting CPI Based on Convolutional Neural Network and Long Short-Term Memory Network[J]. Computer Engineering and Applications, 2022, 58(9): 256-262.

参考文献

[1] KONTONIKAS A.Inflation and inflation uncertainty in the united kingdom evidence from GARCH modeling[J].Economic Modelling，2003，21（3）：525-543.
[2] JONATHAN W，FORECASTING U S.Inflation by Bayesian model averaging[J].Journal of Forecasting，2009，28（21）：131-144.
[3] CARSTENSEN K，HAGEN J，HOSSFELD O，et al.Money demand stability and inflation prediction in the four largest EMU countries[J].Scottish Journal of Political Economy，2010，56（1）：73-93.
[4] KOOP G，KOROBILIS D.Forecasting inflation using dynamic model averaging[J].International Economic Review，2012，53（3）：11-23.
[5] AHMAR A S，ACHMAD G S，LISTYORINI T，et al.Implementation of the ARIMA（p，d，q） method to forecasting CPI data using forecast package in R software[J].Journal of Physics Conference Series，2018，1028（1）：012189.
[6] 赵影，周岳.CPI与实体经济状况、货币供应量的关系及对CPI未来走势的预测[J].金融经济，2010（5）：130-131.
ZHAO Y，ZHOU Y.The relationship between CPI and economic conditions，money supply and the forecast of the future trend of CPI[J].Financial Economy，2010（5）：130-131.
[7] 雷鹏飞.基于季节性ARIMA模型的中国CPI序列分析与预测[J].统计与决策，2014（14）：32-34.
LEI P F.Analysis and forecast of CPI series in China based on seasonal ARIMA model[J].Statistics & Decision，2014（14）：32-34.
[8] 肖良.基于季节性ARIMA模型的居民消费水平预测[J].统计与决策，2016（8）：83-86.
XIAO L.Household consumption forecast based on seasonal ARIMA model[J].Statistics & Decision，2016（8）：83-36.
[9] CHEN X，RACINE J，SWANSON N R.Semiparametric ARX neural-network models with an application to forecasting inflation[J].IEEE Transactions on Neural Networks，2001，12（4）：674-683.
[10] YE Y，SHAO Y，CHEN W.Comparing inflation forecasts using an ε-wavelet twin support vector regression[J].Journal of Information and Computational Science，2013，10（7）：2041-2049.
[11] SERMPINIS G，STASINAKIS C，THEOFILATOS K，et al.Inflation and unemployment forecasting with genetic support vector regression[J].Journal of Forecasting，2014，33（6）.
[12] QUAN J.Two-step testing procedure for price discovery role of futures prices[J].Journal of Futures Markets，1992，12（2）：139-149.
[13] KAUFMANN R K，ULLMAN B.Oil prices，speculation，and fundamentals：interpreting causal relations among spot and futures prices[J].Energy Economics，2009，31（4）：550-558.
[14] WONG W K.Market efficiency of oil spot and futures：a mean-variance and stochastic dominance approach[J].Energy Economics，2010，32（5）：979-986.
[15] BIGMAN D，GOLDFARB D，SCHECHTMAN E.Futures market efficiency and the time content of the information sets[J].Journal of Futures Markets，2010，3（3）：321-334.
[16] BAILLIE R T，MYERS R J.Bivariate GARCH estimation of the optimal commodity futures hedge[J].Journal of Applied Econometrics，2010，6（3）：31-34.
[17] FOSTER A J.Price discovery in oil markets：a time varying analysis of the 1990-91 gulf conflict[J].Energy Economics，1996，18（3）：231-246.
[18] 周舟，成思危.沪深300股指期货市场中的宏观经济信息发布与价格发现[J].系统工程理论与实践，2013，33（12）：3045-3053.
ZHOU Z，CHENG S W.Macroeconomic announcements and price discovery in the CSl 300 stock index futures market[J].Systems Engineering Theory & Practice，2013，33（12）：3045-3053.
[19] 王楠，汪琛德.郑商所农产品期货对通货膨胀的预警作用——基于VAR模型的分析[J].系统管理学报，2015，24（6）：854-858.
WANG N，WANG C D.The warning function of future prices of agricultural products in ZCE to the inflation of China：a VAR analysis[J].Journal of Systems & Management，2015，24（6）：854-858.
[20] GERS F A，CUMMINS F.Learning to forget：continual prediction with LSTM[J].Neural Computation，2000，12（10）：2451-2471.
[21] DENG L，PLATT J.Ensemble deep learning for speech recognition[C]//Proceedings of the Annual Conference of the International Speech Communication Association.New York：IEEE，2014：1915-1919.
[22] BAE S，CHOI I，KIM N.Acoustic scene classification using parallel combination of LSTM and CNN[J].IEEE Signal Processing Magazine，2015，32（3）：16-34.
[23] TARA N，ORIOL V，ANDREW S，et al.Convolutional，long short-term memory，fully connected deep neural networks[C]//International Conference on Acoustics，Speech and Signal Processing.Brisbane，Australia：IEEE，2015：4580-4584.
[24] NG Y H，HAUSKNECHT M，VIJAYANARASIMHAN S，et al.Beyond short snippets：deep networks for video classification[C]//Conference on Computer Vision and Pattern Recognition.Boston，USA：IEEE，2015：4694-4702.