A novel cluster HAR-type model for forecasting realized volatility

Yao, Xingzhi; Izzeldin, Marwan; Li, Zhenxiong

doi:10.1016/j.ijforecast.2019.04.017

Cited by 13 publications

(3 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The goal of this paper is to provide evidence of the superior forecasting performances of the modified HAR-RV models compared with the general approaches. The DM test proposed by Diebold and Mariano (1995) is an appropriate method for one-to-one comparative evaluation and has been widely used by a large number of studies on volatility forecasting (see, e.g., Asai et al, 2020;Chortareas et al, 2011;Kang & Yoon, 2013; E. M. H. Lin et al, 2012;Pan et al, 2017Pan et al, , 2019Qu et al, 2018;Sévi, 2014;Shen & Ritter, 2016;Yao et al, 2019;.…”

Section: Forecasting Evaluationmentioning

confidence: 99%

Forecasting realized volatility of Chinese stock market: A simple but efficient truncated approach

Wen

Zhang

et al. 2021

Journal of Forecasting

View full text Add to dashboard Cite

In this study, we propose a new family of the heterogeneous autoregressive realized volatility (HAR-RV) models by considering truncated methods for predicting the RV in China's stock market. By adopting three types of critical values to recognize extremely large values of RV, we show that the modified models are simple but efficient to consistently deliver stronger in-sample and out-of-sample forecasting performances than those of existing methods. Models that take truncated approaches into account can generate substantial economic gains in applications. We further provide evidence that the superiority of our proposed models is derived from the reduced variance of the measurement errors during days including truncated RVs. Additionally, the improved performances of the modified models still hold after considering the effects of jump components and leverage, as well as a wide range of extensions and robustness analyses. K E Y W O R D SChina's stock market, HAR-RV model, truncated method, volatility forecasting | INTRODUCTIONVolatility and accurate forecasts of volatility in the stock market are crucial for asset pricing and risk management. Given the helpful role of high-frequency data in forecasting volatility, realized volatility (RV) based on intraday data, as pioneered by Andersen and Bollerslev (1998), receives much attention with regard to modeling and predicting financial market volatility (see, e.g.,

show abstract

Section: Forecasting Evaluationmentioning

confidence: 99%

Forecasting realized volatility of Chinese stock market: A simple but efficient truncated approach

Wen

Zhang

et al. 2021

Journal of Forecasting

View full text Add to dashboard Cite

show abstract

“…Long-term memory model can only describe the long-term memory. High frequency prediction models of heterogeneous autoregressive (HAR) volatility and low frequency prediction models are suitable for short-term volatility prediction, and the mixing frequency GARCH model is better than other models in predicting medium and long-term volatilities [37,38], but it mainly focuses on the identification of volatility-influencing factors and simplified description of their action mechanisms. In addition, the heterogeneous autoregressive realized volatility model (HAR family model) proposed by [19] is a simple volatility long memory model.…”

Section: Selection Of Jump Diffusion Volatility Prediction Modelmentioning

confidence: 99%

Jump Aggregation, Volatility Prediction, and Nonlinear Estimation of Banks’ Sustainability Risk

et al. 2020

View full text Add to dashboard Cite

Extreme financial events usually lead to sharp jumps in stock prices and volatilities. In addition, jump clustering and stock price correlations contribute to the risk amplification acceleration mechanism during the crisis. In this paper, four Jump-GARCH models are used to forecast the jump diffusion volatility, which is used as the risk factor. The linear and asymmetric nonlinear effects are considered, and the value at risk of banks is estimated by support vector quantile regression. There are three main findings. First, in terms of the volatility process of bank stock price, the Jump Diffusion GARCH model is better than the Continuous Diffusion GARCH model, and the discrete jump volatility is significant. Secondly, due to the difference of the sensitivity of abnormal information shock, the jump behavior of bank stock price is heterogeneous. Moreover, CJ-GARCH models are suitable for most banks, while ARJI-R2-GARCH models are more suitable for small and medium sized banks. Thirdly, based on the jump diffusion volatility information, the performance of the support vector quantile regression is better than that of the parametric quantile regression and nonparametric quantile regression.

show abstract

“…Wang et al (2017) propose the time‐varying parameters of HAR model with the regression coefficient being assumed to be a function of expression variables and estimated via a local linear regression. Yao et al (2019) constructed a cluster HAR‐type model based on the cluster group lasso method. It is worth noting that under the low‐frequency daily data situations, these existing models enjoy good forecasting performance on the daily and low‐frequency return volatility, whereas they may not be well suitable for the problem of forecasting volatility based on high‐frequency data that usually exhibit high nonlinear pattern and a significant departure from normality.…”

Section: Introductionmentioning

confidence: 99%

Forecasting the high‐frequency volatility based on the LSTM‐HIT model

Liu,

Zhuang,

Wang

2024

Journal of Forecasting

View full text Add to dashboard Cite

Volatility forecasting from high‐frequency data plays a crucial role in many financial fields, such as risk management, option pricing, and portfolio management. Many existing statistical models could better describe and forecast the characteristics of volatility, whereas they do not simultaneously account for the long‐term memory of volatility, the nonlinear characteristics of high‐frequency data, and technical index information during the modeling phase. The purpose of this paper is to use the prediction advantage of deep learning long short‐term memory (LSTM) model to predict the volatility fusing three classes of information, that is, high frequency realized volatility (H), technical indicators (I), and the parameters of generalized autoregression conditional heteroskedasticity(GARCH), heterogeneous autoregressive (HAR), and c, resulting in a novel LSTM‐HIT model to forecast realized volatility. We employ the extreme value theory (EVT) of a semiparametric method to estimate the quantile of standardized return and construct the LSTM‐HIT‐EVT model to forecast the value at risk (VaR). Empirical results show that the LSTM‐HIT model provides the most accurate volatility forecast among the various considered models and that the LSTM‐HIT‐EVT model yields forecasts more accurate than other VaR models.

show abstract

A novel cluster HAR-type model for forecasting realized volatility

Cited by 13 publications

References 32 publications

Forecasting realized volatility of Chinese stock market: A simple but efficient truncated approach

Forecasting realized volatility of Chinese stock market: A simple but efficient truncated approach

Jump Aggregation, Volatility Prediction, and Nonlinear Estimation of Banks’ Sustainability Risk

Forecasting the high‐frequency volatility based on the LSTM‐HIT model

Contact Info

Product

Resources

About