文献类型：期刊文献

英文题名：Short term prediction of extreme returns based on the recurrence interval analysis

作者：Jiang, Zhi-Qiang[1,2,5,6];Wang, Gang-Jin[3,4,5,6];Canabarro, Askery[5,6,7,8];Podobnik, Boris[9,10,11];Xie, Chi[3,4];Stanley, H. Eugene[5,6];Zhou, Wei-Xing[1,2]

机构：[1]East China Univ Sci & Technol, Dept Finance, Shanghai 200237, Peoples R China;[2]East China Univ Sci & Technol, Res Ctr Econophys, Shanghai 200237, Peoples R China;[3]Hunan Univ, Business Sch, Changsha, Hunan, Peoples R China;[4]Hunan Univ, Ctr Finance & Investment Management, Changsha, Hunan, Peoples R China;[5]Boston Univ, Dept Phys, Boston, MA 02215 USA;[6]Boston Univ, Ctr Polymer Studies, Boston, MA 02215 USA;[7]Univ Fed Alagoas, Grp Fis Mat Condensada, Nucleo Ciencias Exatas, Campus Arapiraca, BR-57309005 Arapiraca, AL, Brazil;[8]Univ Fed Rio Grande do Norte, Int Inst Phys, BR-59070405 Natal, RN, Brazil;[9]Univ Rijeka, Fac Civil Engn, Rijeka 51000, Croatia;[10]Zagreb Sch Econ & Management, Zagreb 10000, Croatia;[11]Luxembourg Sch Business, Luxembourg, Luxembourg

年份：2018

卷号：18

期号：3

起止页码：353

外文期刊名：QUANTITATIVE FINANCE

收录：;WOS:【SSCI(收录号:WOS:000424952100003)，SCI-EXPANDED(收录号:WOS:000424952100003)】；

基金：Z.-Q.J. and W.-X.Z. acknowledge support from the National Natural Science Foundation of China (71131007 and 71532009), China Scholarship Council (201406745014) and the Fundamental Research Funds for the Central Universities (222201718006). G.-J.W. and C.X. acknowledge support from the National Natural Science Foundation of China (71501066, 71373072, and 71521061). A.C. acknowledges the support from Brazilian agencies FAPEAL (PPP 20110902-011-0025-0069/60030-733/2011) and CNPq (PDE 20736012014-6, Universal 423713/2016-7). H.E.S. was supported by NSF (Grants CMMI 1125290, PHY 1505000, and CHE-1213217) and by DOE Contract (DE-AC07-05Id14517).

语种：英文

外文关键词：Extreme return; Risk estimation; Recurrence interval; Return forecasting; Hazard probability; G130; G110

摘要：Being able to predict the occurrence of extreme returns is important in financial risk management. Using the distribution of recurrence intervalsthe waiting time between consecutive extremeswe show that these extreme returns are predictable in the short term. Examining a range of different types of returns and thresholds we find that recurrence intervals follow a q-exponential distribution, which we then use to theoretically derive the hazard probability http://www.w3.org/1999/xlink. Maximizing the usefulness of extreme forecasts to define an optimized hazard threshold, we indicate a financial extreme occurring within the next day when the hazard probability is greater than the optimized threshold. Both in-sample tests and out-of-sample predictions indicate that these forecasts are more accurate than a benchmark that ignores the predictive signals. This recurrence interval finding deepens our understanding of reoccurring extreme returns and can be applied to forecast extremes in risk management.