Estimating and Forecasting Volatility of the Stock Indices Using Conditional Autoregressive Range (Carr) Model (original) (raw)
Related papers
A Range-Based GARCH Model for Forecasting Volatility
A new variant of the ARCH class of models for forecasting the conditional variance, to be called the Generalized AutoRegressive Conditional Heteroskedasticity Parkinson Range (GARCH-PARK-R) model, is proposed. The GARCH-PARK-R model, utilizing the extreme values, is a good alternative to the “realized volatility” model which requires a large amount of intra-daily data that remain relatively costly and are not readily available. The estimates of the GARCH-PARK-R model are derived using the Quasi-Maximum Likelihood Estimation (QMLE). The results suggest that the GARCHPARK- R model is a good middle ground between intra-daily models, such as the realized volatility, and inter-daily models, such as the ARCH class. The forecasting performance of the models is evaluated using the daily Philippine Peso-U.S. Dollar exchange rate from January 1997 to December 2003.
Financial volatility forecasting with range-based autoregressive volatility model
Finance Research Letters, 2011
The classical volatility models, such as GARCH, are return-based models, which are constructed with the data of closing prices. It might neglect the important intraday information of the price movement, and will lead to loss of information and efficiency. This study introduces and extends the range-based autoregressive volatility model to make up for these weaknesses. The empirical results consistently show that the new model successfully captures the dynamics of the volatility and gains good performance relative to GARCH model.
Econometric Modeling: Capital Markets - Forecasting eJournal, 2021
Modelling volatility has become increasingly important in recent times for its diverse implications. The main purpose of this paper is to examine the performance of volatility modelling using different models and their forecasting accuracy for the returns of Dhaka Stock Exchange (DSE) under different error distribution assumptions. Using the daily closing price of DSE from the period 27 January 2013 to 06 November 2017, this analysis has been done using Generalized Autoregressive Conditional Heteroscedastic (GARCH), Asymmetric Power Autoregressive Conditional Heteroscedastic (APARCH), Exponential Generalized Autoregressive Conditional Heteroscedastic (EGARCH), Threshold Generalized Autoregressive Conditional Heteroscedastic (TGARCH) and Integrated Generalized Autoregressive Conditional Heteroscedastic (IGARCH) models under both normal and student’s t error distribution. The study finds that ARMA (1,1)- TGARCH (1,1) is the most appropriate model for in-sample estimation accuracy unde...
2004
A new variant of the ARCH class of models for forecasting the conditional variance, to be called the Generalized AutoRegressive Conditional Heteroskedasticity Parkinson Range (GARCH-PARK-R) Model, is proposed. The GARCH-PARK-R model, utilizing the extreme values, is a good alternative to the Realized Volatility that requires a large amount of intra-daily data. The estimates of the GARCH-PARK-R model are derived using the Quasi-Maximum Likelihood Estimation (QMLE). The Parkinson Range is also used to evaluate the out-ofsample forecasting performance of 68 ARCH models using the inter-daily Philippine Peso-U.S. Dollar exchange rate.
Comparative Study of Volatility Forecasting Models: The Case
This paper aims to investigate the effectiveness of four volatility forecasting models, i.e. Exponential Weighted Moving Average (EWMA), Autoregressive Integrated Moving Average (ARIMA) and Generalized Auto-Regressive Conditional Heteroscedastic (GARCH), in four stock markets Indonesia, Malaysia, Japan and Hong Kong. Using monthly closing stock index prices collected from 1st January 1998 to 31st December 2015 for the four selected countries, results obtained confirm that volatility in developed markets is not necessarily always lower than the volatility in emerging markets. Among all the three models, GARCH (1, 1) model is found to be the best forecasting model for stock markets in Malaysia, Indonesia, and Japan, while EWMA model is found to be the best forecasting model for Hong Kong stock market. The outperformance of GARCH (1, 1) found supports again what is found in Minkah (2007).
Journal of Business Studies, 2022
The aim of this research is to identify the best-fitted model(s) for estimating and forecasting the return volatility of the Chittagong Stock Exchange (CSE) in Bangladesh. Methodology: The study analyzes the returns of the Chittagong Stock Exchange's (CSE) daily Selective Categories Index (CSCX) from February 4, 2013 to December 31, 2021 (as a full sample) and from July 1, 2021 to December 30, 2021 (for forecasting). The researcher used GARCH family approaches considering different error distributions, to find the well-suited model(s) for the CSCX index. The researchers used ARMA to develop the mean equation based on two popular model selection criteria: Schwarz's (1978) Bayesian information criterion (SBIC) and Akaike's (1974) information criterion (AIC). The data has been analyzed using the application software E-Views 10. Findings: The ARMA (0,1) has been adopted as the mean equation for GARCH specifications. Under all three types of error distributions, the ARCH and GARCH terms, along with the leverage terms of asymmetric models, were found to be statistically significant in all the accepted combinations of the model. The models GARCH (1,2), TGARCH (1,2), and PARCH (1,2) under generalized error distributions and EGARCH (2,1) under Student"s t error distributions have been selected as the bestfitted models for estimation. Whereas, based on root mean square error (RMSE), mean absolute error (MAE), mean absolute percent error (MAPE), and theil inequality (TI), EGARCH (1, 2), TGARCH (1, 2), and PARCH (1, 2) under generalized error distributions, and GARCH (1, 2) under student"s t error distributions and normal error distributions are found to have superior out-of-sample forecasting abilities. Practical implications and originality: This is an original research work that will help the investors and other stakeholders of the Bangladeshi stock market to estimate and forecast market volatility more efficiently. Limitations: Due to its extensive features, this study was unable to incorporate a few additional ARCH and GARCH models.
Volatility forecasting using threshold heteroskedastic models of the intra-day range
Computational Statistics & Data Analysis, 2008
An effective approach for forecasting return volatility via threshold nonlinear heteroskedastic models of the daily asset price range is provided. The return is defined as the difference between the highest and lowest log intra-day asset price. A general model specification is proposed, allowing the intra-day high-low price range to depend nonlinearly on past information, or an exogenous variable such as US market information. The model captures aspects such as sign or size asymmetry and heteroskedasticity, which are commonly observed in financial markets. The focus is on parameter estimation, inference and volatility forecasting in a Bayesian framework. An MCMC sampling scheme is employed for estimation and shown to work well in simulation experiments. Finally, competing range-based and return-based heteroskedastic models are compared via out-of-sample forecast performance. Applied to six international financial market indices, the range-based threshold heteroskedastic models are well supported by the data in terms of finding significant threshold nonlinearity, diagnostic checking and volatility forecast performance under various volatility proxies.
GARCH-based Volatility Forecasts for Market Volatility Indices
2002
Volatility forecasting is one of the main issues in the financial econometrics literature. Volatility measures may be derived from statistical models for conditional variance, or from option prices. In recent times, indices have been suggested which summarize the implied volatility of widely traded market index options. One such index is the so-called VXN, an average of 30-day ahead implied volatilities of the options written on the NASDAQ-100 Index. In this paper we show how forecasts obtained with traditional GARCH-type models can be used to forecast the volatility index VXN.
Forecasting Equity Index Volatility: Empirical Evidence from Japan, UK and USA Data
Financial Risk and Management Reviews
Using non-linear models to forecast volatility for three equity index samples, this study examines weekly returns of three indices; Dow Jones Industrial index, FTSE 100 index, and Nikkei 225 index. The sample covers a twenty year sample period. The study employs an in sample and out of sample volatility forecast using standard symmetric loss functions in order to identify an appropriate model that best forecast volatility. Using the mean error (ME), root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE), the study finds the EGARCH model to outperform the ARCH, and GARCH model in forecasting volatility. Contribution/Originality: This is among the first studies that found EGARCH model to outperform the ARCH, and GARCH model in forecasting volatility using a combination of Japan, UK and US data.