时变参数模型的最优滚动窗宽选择标准及应用

doi:10.16381/j.cnki.issn1003-207x.2018.08.003

摘要/Abstract

摘要： 宏观经济领域中存在严重的结构突变性，模型估计量的优劣对估计样本规模是敏感的。本文针对时变参数模型，建立了滚动窗宽选择标准，通过最小化估计量的近似二次损失函数及最大化各子样本估计量间的曼哈顿距离选择窗宽大小，权衡了模型估计量的准确性和时变性两个相悖目标。蒙特卡罗模拟实验表明，本文所提出的方法在各种结构突变情形下均适用，能够应用于线性关系和非线性关系的时变参数模型中，且均具有稳健性。将该方法应用于我国金融网络的结构突变识别过程，显著改善了传统窗宽选择方法的结果。

关键词: 滚动窗宽选择, 时变参数模型, Bootstrap估计, 金融网络

Abstract: There are serious structural changes in the macroeconomic field. The performance of the model estimators is sensitive to the choice of estimation sample size, while methods to select the window size in rolling time-varying parameter model have received little attention.In this paper, a new approach is developed to select the rolling bandwidth for capturing the time-varying parameter in models with potential breaks. More specifically, the function forms are unknown, which can be set as a single-index semi-parametric model that can capture the linear or nonlinear relationship between variables, also can be extended to the linear or generalized linear regression model where only need to use the corresponding model estimation method.
Our new approach, to balance the accuracy and the time-varying objectives of the model estimators, solves the multi-objective optimization problem that minimizing bootstrap approximation quadratic loss function of model estimators and maximizing the Manhattan distance between the sub-sample estimators. Monte Carlo simulations show that using the window size selected by our procedure can significantly improve upon the performance of the model estimators. And also our method is applicable to all kinds of structural changes and time-varying parameter models of linear and nonlinear relations, not sensitive to the parameters choice in the same data generation process.When applied to capture the structural changes of China's financial network, 30 financial institutions, from 16 October, 2010 to 26 September, 2015, are included. Our results suggest that our procedure can capture the structural changes of the financial system, also significantly improve upon the performance of the financial network model estimators compared to traditional methods which just according to the subjective intention and forecasting performance. Our research and conclusions are helpful for the optimization and application of time-varying parameter model, and have important theoretical value and practical significance.

Key words: rolling bandwidth selection, time-varying parameter model, bootstrap estimation, financial network

中图分类号:

O212
F830.99

傅强, 张兴敏. 时变参数模型的最优滚动窗宽选择标准及应用[J]. 中国管理科学, 2018, 26(8): 20-30.

FU Qiang, ZHANG Xing-min. Optimal Rolling Window Selection for Time-Varying Parameter Model and Its Application[J]. Chinese Journal of Management Science, 2018, 26(8): 20-30.

参考文献

[1] Granger C W J. Can we improve the perceived quality of economic forecasts?[J].Journal of Applied Econometrics, 1996, 11(5):455-473.

[2] Nyakabawo W, Miller S M, Balcilar M, et al. Temporal causality between house prices and output in the US:A Bootstrap rollingwindow approach[J]. The North American Journal of Economics and Finance, 2015, 33:55-73.

[3] Straetmans S, Chaudhry S M. Tail risk and systemic risk of US and Eurozone financial institutions in the wake of the global financial crisis[J]. Journal of International Money and Finance, 2015, 58:191-223.

[4] Betz F, Hautsch N, Peltonen T A, et al. Systemic risk spillovers in the European banking and sovereign network[J]. Journal of Financial Stability, 2015,25:206-224.

[5] Hautsch N, Schaumburg J, Schienle M. Financial network systemic risk contributions[J]. Review of Finance,2015,19(2):685-738.

[6] Bessler W, Kurmann P, Nohel T. Time varying systematic and idiosyncratic risk exposures of US bank holding companies[J]. Journal of International Financial Markets Institutions & Money, 2015, 35:45-68.

[7] Härdle W K, Wang Weining, Yu Lining. TENET:Tail-event driven NETwork risk[J]. Journal of Econometrics, 2016, 192(2):499-513.

[8] Caggiano G, Calice P, Leonida L, et al. Comparing logit-based early warning systems:Does the duration of systemic banking crises matter?[J]. Journal of Empirical Finance, 2016, 37:104-116.

[9] Shahbaz M, Shahzad S J H, Jammazi R. Nexus between US energy sources and economic activity:Time-frequency and bootstrap rolling window causality analysis[R]. Munich Personal RePEC Archive, 2016.

[10] Stock J H, Watson M W. Evidence on structural instability in macroeconomic time series relations[J]. Journal of Business & Economic Statistics, 1996, 14(1):11-30.

[11] Stock J H, Watson M W. Forecasting output and inflation:The role of asset prices[J]. Journal of Economic Literature, 2003, 41(3):788-829.

[12] Stock J H, Watson M W. Why has US inflation become harder to forecast?[J]. Journal of Money, Credit and Banking, 2007, 39(S1):3-33.

[13] Pesaran M H, Timmermann A. Selection of estimation window in the presence of breaks[J]. Journal of Econometrics, 2007, 137(1):134-161.

[14] Pesaran M H, Pick A, Pranovich M. Optimal forecasts in the presence of structural breaks[J]. Journal of Econometrics, 2013, 177(2):134-152.

[15] Giraitis L, Kapetanios G, Price S. Adaptive forecasting in the presence of recent and ongoing structural change[J]. Journal of Econometrics, 2013, 177(2):153-170.

[16] Inoue A, Jin Lu, Rossi B. Rolling window selection for out-of-sample forecasting with time-varying parameters[J]. Journal of Econometrics, 2017, 196(1):55-67.

[17] Khediri K B, Charfeddine L. Evolving efficiency of spot and futures energy markets:A rolling sample approach[J]. Journal of Behavioral & Experimental Finance, 2015, 6:67-79.

[18] López-Espinosa G, Moreno A, Rubia A, et al. Systemic risk and asymmetric responses in the financial industry[J]. Journal of Banking & Finance, 2015, 58:471-485.

[19] Charfeddine L, Benlagha N. A time-varying copula approach for modelling dependency:New evidence from commodity and stock markets[J]. Journal of Multinational Financial Management, 2016,37-38:168-189.

[20] Pesaran M H, Timmermann A. Small sample properties of forecasts from auto-regressive models under structural breaks[J]. Journal of Econometrics, 2005, 129(1):183-217.

[21] Anagnostidis P, Varsakelis C, Emmanouilides C J. Has the 2008 financial crisis affected stock market efficiency? The case of Eurozone[J]. Physica A:Statistical Mechanics and its Applications, 2016, 447:116-128.

[22] Härdle W, Mammen E. Testing parametric versus nonparametric regression[J]. Annals of Statistics, 1993, 21:1926-1947.

[23] Carroll R J, Fan J, Gijbels I, et al. Generalized partially linear single-index models[J]. Journal of the American Statistical Association, 1997, 92(438):477-489.

[24] Zhang Riquan, Huang Zhensheng, Lv Yazhao. Statistical inference for the index parameter in single-index models[J]. Journal of Multivariate Analysis, 2010, 101(4):1026-1041.

[25] Chaudhuri P, Doksum K, Samarov A. On average derivative quantile regression[J]. The Annals of Statistics, 1997, 25(2):715-744.

[26] Murata T, Ishibuchi H, Tanaka H. Multi-objective genetic algorithm and its applications to flowshop scheduling[J]. IEEE Transactions on Systems Man & Cybernetics Part C, 1996, 30(4):957-968.

[27] Hall P. Using the bootstrap to estimate mean squared error and select smoothing parameter in nonparametric problems[J]. Journal of Multivariate Analysis, 1990, 32(2):177-203.

[28] Lombardla M J, Gonzalez-Manteiga W, Prada-Sánchez J M. Bootstrapping the Chambers-Dunstan estimate of a finite population distribution function[J]. Journal of Statistical Planning and Inference, 2003, 116(2):367-388.

[29] Manteiga W G, Miranda M D M, González A P. The choice of smoothing parameter in nonparametric regression through Wild Bootstrap[J]. Computational Statistics & Data Analysis, 2004, 47(3):487-515.

[30] González-Manteiga W, Lombardía M J, Molina I, et al. Estimation of the mean squared error of predictors of small area linear parameters under a logistic mixed model[J]. Computational Statistics & Data Analysis, 2007, 51(5):2720-2733.

[31] Marchetti S, Tzavidis N, Pratesi M. Non-parametric Bootstrap mean squared error estimation for M-quantile estimators of small area averages, quantiles and poverty indicators[J]. Computational Statistics & Data Analysis, 2012, 56(10):2889-2902.

[32] Paltalidis N, Gounopoulos D, Kizys R, et al. Transmission channels of systemic risk and contagion in the European financial network[J]. Journal of Banking & Finance, 2015, 61(S1):S36-S52.

[33] 陈守东, 王妍. 我国金融机构的系统性金融风险评估——基于极端分位数回归技术的风险度量[J]. 中国管理科学, 2014, 22(7):10-17.

[34] 高大良, 刘志峰, 杨晓光. 投资者情绪、平均相关性与股市收益[J]. 中国管理科学, 2015, 23(2):10-20.

[35] 张宗新, 王海亮. 投资者情绪、主观信念调整与市场波动[J]. 金融研究, 2013,(4):142-155.

[36] Sedunov J. What is the systemic risk exposure of financial institutions?[J]. Journal of Financial Stability, 2016, 24:71-87.

[37] Black L, Correa R, Huang Xin, et al. The systemic risk of European banks during the financial and sovereign debt crises[J]. Journal of Banking & Finance, 2016, 63:107-125.