Computational Statistics

, Volume 30, Issue 3, pp 821–843

Recurrent support vector regression for a non-linear ARMA model with applications to forecasting financial returns

Original Paper

DOI: 10.1007/s00180-014-0543-9

Cite this article as:
Chen, S., Jeong, K. & Härdle, W.K. Comput Stat (2015) 30: 821. doi:10.1007/s00180-014-0543-9
  • 257 Downloads

Abstract

Motivated by recurrent neural networks, this paper proposes a recurrent support vector regression (SVR) procedure to forecast nonlinear ARMA model based simulated data and real data of financial returns. The forecasting ability of the recurrent SVR based ARMA model is compared with five competing models (random walk, threshold ARMA model, MLE based ARMA model, recurrent artificial neural network based ARMA model and feed-forward SVR based ARMA model) by using two forecasting accuracy evaluation metrics (NSME and sign) and robust Diebold–Mariano test. The results reveal that for one-step-ahead forecasting, the recurrent SVR model is consistently better than the benchmark models in forecasting both the magnitude and turning points, and statistically improves the forecasting performance as opposed to the usual feed-forward SVR.

Keywords

Recurrent support vector regression Non-linear ARMA   Financial forecasting 

JEL Classification

C45 C53 F37 F47 G17 

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  1. 1.China Center for Economic Studies, School of EconomicsFudan UniversityShanghaiChina
  2. 2.School of Economics and TradeKyungpook National UniversityDaeguKorea
  3. 3.Center for Applied Statistics and EconomicsHumboldt-Universität zu BerlinBerlinGermany
  4. 4.Lee Kong Chian School of BusinessSingapore Management UniversitySingaporeSingapore

Personalised recommendations