A novel hybrid framework for forecasting stock indices based on the nonlinear time series models

Hasnain Iftikhar (), Faridoon Khan (), Elías A. Torres Armas (), Paulo Canas Rodrigues () and Javier Linkolk López-Gonzales ()
Additional contact information
Hasnain Iftikhar: Quaid-i-Azam University
Faridoon Khan: Pakistan Institute of Development Economics
Elías A. Torres Armas: Universidad Nacional Toribio Rodríguez de Mendoza
Paulo Canas Rodrigues: Federal University of Bahia
Javier Linkolk López-Gonzales: Universidad Peruana Unión

Computational Statistics, 2025, vol. 40, issue 8, No 5, 4163-4186

Abstract: Abstract This study presents a new hybrid forecasting system to enhance the accuracy and efficiency of predicting stock market trends. To do this, the proposed involves several steps. Firstly, the closed stock index price time series is preprocessed to address missing values, variance stabilization, nonnormality, and nonstationarity. Second, the stock index closing prices are processed and filtered into a nonlinear long-term trend series and a stochastic series using three proposed filters and a benchmark filter. Third, the filtered series are estimated using the nonlinear and neural network autoregressive models. Fourth, the residual from both the fitted models is extracted, and a new series is obtained. The new residual series is forecasted using the autoregressive condition heterogeneity model. Finally, the forecasts from each model are combined to get the final estimates. The results indicate that the proposed final hybrid model produced the most accurate and efficient comparison with the baseline models.

Keywords: Complex G7 stock markets; Daily closing stock index forecasting; Regression type filters; Nonlinear time series models; Hybrid forecasting system (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s00180-025-01614-5 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:compst:v:40:y:2025:i:8:d:10.1007_s00180-025-01614-5

Ordering information: This journal article can be ordered from
http://www.springer.com/statistics/journal/180/PS2

DOI: 10.1007/s00180-025-01614-5

Access Statistics for this article

Computational Statistics is currently edited by Wataru Sakamoto, Ricardo Cao and Jürgen Symanzik

More articles in Computational Statistics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().