Forecasting the Volatility of Stock Market Index Using the Hybrid Models with Google Domestic Trends

Monghwan Seo, Sung chul Lee, Geonwoo Kim

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In order to improve the forecasting accuracy of the volatilities of the markets, we propose the hybrid models based on artificial neural networks with multi-hidden layers in this paper. Specifically, the hybrid models are built using the estimated volatilities obtained from GARCH family models and Google domestic trends (GDTs) as input variables. We further carry out many experiments varying the number of layers and activation functions to obtain the accurate hybrid model for forecasting volatility. The proposed models are applied to forecast weekly and monthly volatilities of S&P 500 index to verify their accuracy. The performance comparison results show that the hybrid models with GDTs outperform clearly the predicted results with GARCH family models and the hybrid models without GDTs in forecasting the volatility of actual market. We also provide the experiment results with graphs to illustrate the efficiency of models.

Original languageEnglish
Article number1950006
JournalFluctuation and Noise Letters
Volume18
Issue number1
DOIs
Publication statusPublished - 2019 Mar 1

Fingerprint

volatility
Hybrid Model
Stock Market
forecasting
Volatility
Forecasting
trends
Generalized Autoregressive Conditional Heteroscedasticity
Volatility Forecasting
Activation Function
Performance Comparison
Comparison Result
Model
Experiment
Artificial Neural Network
Forecast
Trends
Model-based
Verify
Graph in graph theory

All Science Journal Classification (ASJC) codes

  • Mathematics(all)
  • Physics and Astronomy(all)

Cite this

@article{a9f5ed0e2fa346ccaff58bff319f9ee3,
title = "Forecasting the Volatility of Stock Market Index Using the Hybrid Models with Google Domestic Trends",
abstract = "In order to improve the forecasting accuracy of the volatilities of the markets, we propose the hybrid models based on artificial neural networks with multi-hidden layers in this paper. Specifically, the hybrid models are built using the estimated volatilities obtained from GARCH family models and Google domestic trends (GDTs) as input variables. We further carry out many experiments varying the number of layers and activation functions to obtain the accurate hybrid model for forecasting volatility. The proposed models are applied to forecast weekly and monthly volatilities of S&P 500 index to verify their accuracy. The performance comparison results show that the hybrid models with GDTs outperform clearly the predicted results with GARCH family models and the hybrid models without GDTs in forecasting the volatility of actual market. We also provide the experiment results with graphs to illustrate the efficiency of models.",
author = "Monghwan Seo and Lee, {Sung chul} and Geonwoo Kim",
year = "2019",
month = "3",
day = "1",
doi = "10.1142/S0219477519500068",
language = "English",
volume = "18",
journal = "Fluctuation and Noise Letters",
issn = "0219-4775",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "1",

}

Forecasting the Volatility of Stock Market Index Using the Hybrid Models with Google Domestic Trends. / Seo, Monghwan; Lee, Sung chul; Kim, Geonwoo.

In: Fluctuation and Noise Letters, Vol. 18, No. 1, 1950006, 01.03.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Forecasting the Volatility of Stock Market Index Using the Hybrid Models with Google Domestic Trends

AU - Seo, Monghwan

AU - Lee, Sung chul

AU - Kim, Geonwoo

PY - 2019/3/1

Y1 - 2019/3/1

N2 - In order to improve the forecasting accuracy of the volatilities of the markets, we propose the hybrid models based on artificial neural networks with multi-hidden layers in this paper. Specifically, the hybrid models are built using the estimated volatilities obtained from GARCH family models and Google domestic trends (GDTs) as input variables. We further carry out many experiments varying the number of layers and activation functions to obtain the accurate hybrid model for forecasting volatility. The proposed models are applied to forecast weekly and monthly volatilities of S&P 500 index to verify their accuracy. The performance comparison results show that the hybrid models with GDTs outperform clearly the predicted results with GARCH family models and the hybrid models without GDTs in forecasting the volatility of actual market. We also provide the experiment results with graphs to illustrate the efficiency of models.

AB - In order to improve the forecasting accuracy of the volatilities of the markets, we propose the hybrid models based on artificial neural networks with multi-hidden layers in this paper. Specifically, the hybrid models are built using the estimated volatilities obtained from GARCH family models and Google domestic trends (GDTs) as input variables. We further carry out many experiments varying the number of layers and activation functions to obtain the accurate hybrid model for forecasting volatility. The proposed models are applied to forecast weekly and monthly volatilities of S&P 500 index to verify their accuracy. The performance comparison results show that the hybrid models with GDTs outperform clearly the predicted results with GARCH family models and the hybrid models without GDTs in forecasting the volatility of actual market. We also provide the experiment results with graphs to illustrate the efficiency of models.

UR - http://www.scopus.com/inward/record.url?scp=85056591396&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056591396&partnerID=8YFLogxK

U2 - 10.1142/S0219477519500068

DO - 10.1142/S0219477519500068

M3 - Article

AN - SCOPUS:85056591396

VL - 18

JO - Fluctuation and Noise Letters

JF - Fluctuation and Noise Letters

SN - 0219-4775

IS - 1

M1 - 1950006

ER -