Asymmetric Generalized Autoregressive Conditional Heteroscedasticity Research Articles

Analysing volatility patterns in emerging markets: symmetric or asymmetric models?

PurposeInvestors aim for returns when investing in stocks, making return volatility a crucial concern. This study compares symmetric and asymmetric GARCH models to forecast volatility in emerging nations like the G4 countries. Accurate volatility forecasting is vital for investors to make well-informed investment decisions, forming the core purpose of this study.Design/methodology/approachFrom January 1993 to May 2021, the study spans four periods, focusing on the global economic crisis of 2008, the Russian crisis of 2015 and the COVID-19 pandemic. Standard generalized autoregressive conditional heteroscedasticity (GARCH), exponential GARCH (E-GARCH) and Glosten-Jagannathan-Runkle GARCH models were employed to analyse the data. Robustness was assessed using the Akaike information criterion, Schwarz information criterion and maximum log-likelihood criteria.FindingsThe study's findings show that the E-GARCH model is the best model for forecasting volatility in emerging nations. This is because the E-GARCH model is able to capture the asymmetric effects of positive and negative shocks on volatility.Originality/valueThis unique study compares symmetric and asymmetric GARCH models for forecasting volatility in emerging nations, a novel approach not explored in prior research. The insights gained can aid investors in constructing more effective risk-adjusted international portfolios, offering a better understanding of stock market volatility to inform strategic investment decisions.

Garch Models Comparison with Symmetric and Asymmetric Process for Univariate Econometric Series

The Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model was modeled using both symmetric and asymmetric processes. Secondary data from January 2005 to December 2021 on the Consumer Price Index, Exchange Rate, Crude Oil Price, and Inflation Rate were used for this study. The research was conducted using the statistical software packages Minitab and E-view. The aforementioned four macroeconomic variables show a tendency for volatility to cluster across time. In both symmetric and asymmetric processes, the volatility condition and leverage impact coefficients were present. By contrasting the symmetric models (ARCH, GARCH, and GARCH-M) and the asymmetric models, the best model was chosen using Akaike Information Criteria (E-GARCH, T-GARCH and APARCH). For the investigated univariate economic variables, the results indicated that the found asymmetric model GARCH models outperformed the symmetric model GARCH models. Therefore, these models can be applied to the forecasting of these series of economic indicators. Models include the Asymmetric E-GARCH (1, 1) Model for Consumer Price Index, Crude Oil Price, and Inflation Rate Series and the Asymmetric T-GARCH (1, 1) Model for Exchange Rate Series.

Linear time-varying regression with copula–DCC–asymmetric–GARCH models for volatility: the co-movement between industrial electricity demand and financial factors

ABSTRACT This paper examines the dependence structure of industrial electricity demand and financial indicators (the Korea Composite Stock Price Index [KOSPI], Korean Securities Dealers Automated Quotations [KOSDAQ], exchange rate, government bonds and exports) using the copula dynamic conditional correlation with symmetric and asymmetric generalized autoregressive conditional heteroscedasticity (GARCH) models to forecast volatility. We investigated symmetric and asymmetric GARCH types, such as the standard, exponential, Glosten–Jagannathan–Runkle and asymmetric power models to fit the marginal distribution. The two types of elliptical copula, Gaussian and Student’s t-distributions, were also considered to investigate the tail dependence between financial and electricity time series. We analysed the monthly log returns for January 2002 to April 2020. The empirical results reveal that the best-fit models for the Akaike information criteria are the asymmetric GARCH models, specifically the exponential-GARCH (E-GARCH). Moreover, the asymmetric GARCH model is superior to the symmetric GARCH in terms of forecast volatility. Extreme tail dependence exists for the KOSDAQ and exports indicators with the electricity demand. The KOSPI, Korea’s primary stock market, is the best-fit financial variable and presents the highest forecasting accuracy.

Volatility of Daily Nepal Stock Exchange (NEPSE) Index Return: A Garch Family Models

The major intend of this study is to investigate the volatility clustering in NEPSE index. To reach the conclusion, 3392 annually observed time series data from 1 June 2006 to 7 April 2021 were obtained from various volume of annual trading report of Nepal Stock Exchange (NEPSE) and website of NEPSE and symmetric Generalized Autoregressive Conditional Heteroscedasticity (GARCH) models––GARCH (1,1), GARCH-M (1,1) and asymmetric GARCH family models––TGARCH (1,1), EGARCH (1,1), and PGARCH (1,1) were employed. The stylized facts confirm that the volatility clustering and leverage effect on the return of NEPSE index are existed. The empirical analysis reveals that the positive correlation between volatility and the expected return of NEPSE index in terms of risk premium and then conditional variance process is persistent. The empirical results also show that the symmetric model is better fitted to full sampled period and asymmetric GARCH family models to before-and after-earthquake sampled period. This study covers the larger dataset which is divided into different episodes with different economic condition of Nepal and thus, it is assumed to be a purely an initial work on Nepalese stock exchange.

Modelling and forecasting of growth rate of new COVID-19 cases in top nine affected countries: Considering conditional variance and asymmetric effect.

COVID-19 pandemic has affected more than a hundred fifty million people and killed over three million people worldwide over the past year. During this period, different forecasting models have tried to forecast time path of COVID-19 pandemic. Unlike the COVID-19 forecasting literature based on Autoregressive Integrated Moving Average (ARIMA) modelling, in this paper new COVID-19 cases were modelled and forecasted by conditional variance and asymmetric effects employing Generalized Autoregressive Conditional Heteroscedasticity (GARCH), Threshold GARCH (TARCH) and Exponential GARCH (EGARCH) models. ARMA, ARMA-GARCH, ARMA-TGARCH and ARMA-EGARCH models were employed for one-day ahead forecasting performance for April, 2021 and three waves of COVID-19 pandemic in nine most affected countries —USA, India, Brazil, France, Russia, UK, Italy, Spain and Germany. Empirical results show that ARMA-GARCH models have better forecast performance than ARMA models by modelling both the conditional heteroskedasticity and the heavy-tailed distributions of the daily growth rate of the new confirmed cases; and asymmetric GARCH models show mixed results in terms of reducing the root mean squared error (RMSE).

Volatility of Stock Prices in Tanzania: Application of Garch Models to Dar Es Salaam Stock Exchange

We use Generalized Autoregressive Conditional Heteroscedasticity (GARCH) models to examine volatility of stock prices for firms listed in the Dar es Salaam Stock Exchange (DSE). In doing so, both symmetric and asymmetric GARCH models are used in this study. The descriptive analysis of the data shows that standard deviation of the series returns is high, indicating a high level of daily fluctuations, and the log value of the mean is close to zero. Our empirical results clearly exhibit evidence of volatility and volatility clustering, a typical feature of financial time series. Moreover, our results indicate that the series are highly leptokurtic, flat tailed and asymmetric consistent with characteristics of financial time series data. Out of all models examined, EGARCH (1,1) and GARCH (1,1) seem to perform plausibly better than others.

Forecasting Volatility of Energy Commodities: Comparison of GARCH Models with Support Vector Regression

We compare the forecasting performance of the generalized autoregressive conditional heteroscedasticity (GARCH) -type models with support vector regression (SVR) for futures contracts of selected energy commodities: Crude oil, natural gas, heating oil, gasoil and gasoline. The GARCH models are commonly used in volatility analysis, while SVR is one of machine learning methods, which have gained attention and interest in recent years. We show that the accuracy of volatility forecasts depends substantially on the applied proxy of volatility. Our study confirms that SVR with properly determined hyperparameters can lead to lower forecasting errors than the GARCH models when the squared daily return is used as the proxy of volatility in an evaluation. Meanwhile, if we apply the Parkinson estimator which is a more accurate approximation of volatility, the results usually favor the GARCH models. Moreover, it is difficult to choose the best model among the GARCH models for all analyzed commodities, however, forecasts based on the asymmetric GARCH models are often the most accurate. While, in the class of the SVR models, the results indicate the forecasting superiority of the SVR model with the linear kernel and 15 lags, which has the lowest mean square error (MSE) and mean absolute error (MAE) among the SVR models in 92% cases.

Modeling Stock Return Data Using Asymmetric Volatility Models: A Performance Comparison Based On the Akaike Information Criterion and Schwarz Criterion

The Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model has been widely used in time series forecasting especially with asymmetric volatility data. As the generalization of autoregressive conditional heteroscedasticity model, GARCH is known to be more flexible to lag structures. Some enhancements of GARCH models were introduced in literatures, among them are Exponential GARCH (EGARCH), Threshold GARCH (TGARCH) and Asymmetric Power GARCH (APGARCH) models. This paper aims to compare the performance of the three enhancements of the asymmetric volatility models by means of applying the three models to estimate real daily stock return volatility data. The presence of leverage effects in empirical series is investigated. Based on the value of Akaike information and Schwarz criterions, the result showed that the best forecasting model for our daily stock return data is the APARCH model.

Analysis of Asymmetric GARCH Volatility Models with Applications to Margin Measurement

We explore properties of asymmetric generalized autoregressive conditional heteroscedasticity (GARCH) models in the threshold GARCH family and propose a more general Spline-GTARCH model, which captures high-frequency return volatility, low-frequency macroeconomic volatility as well as an asymmetric response to past negative news in both autoregressive conditional heteroscedasticity (ARCH) and GARCH terms. Based on maximum likelihood estimation of S\&P 500 returns, S\&P/TSX returns and Monte Carlo numerical example, we find that the proposed more general asymmetric volatility model has better fit, higher persistence of negative news, higher degree of risk aversion and significant effects of macroeconomic variables on the low frequency volatility component. We then apply a variety of volatility models including asymmetric GARCH, GARCH and EWMA in setting initial margin requirements for central clearing counterparties (CCPs).

A behavioral analysis of the volatility of interbank interest rates in developed and emerging countries

Purpose This paper aims to analyse the volatility of the fixed income market from 11 countries (Brazil, Russia, India, China, South Africa, Argentina, Chile, Mexico, USA, Germany and Japan) from January 2000 to December 2011 by examining the interbank interest rates from each market. Design/methodology/approach To the volatility of interest rates returns, the study used models of auto-regressive conditional heteroscedasticity, autoregressive conditional heteroscedasticity (ARCH), generalized autoregressive conditional heteroscedasticity (GARCH), exponential generalized autoregressive conditional heteroscedasticity (EGARCH), threshold generalized autoregressive conditional heteroscedasticity (TGARCH) and periodic generalized autoregressive conditional heteroscedasticity (PGARCH), and a combination of these with autoregressive integrated moving average (ARIMA) models, checking which of these processes were more efficient in capturing volatility of interest rates of each of the sample countries. Findings The results suggest that for most markets, studied volatility is best modelled by asymmetric GARCH processes – in this case the EGARCH – demonstrating that bad news leads to a higher increase in the volatility of these markets than good news. In addition, the causes of increased volatility seem to be more associated with events occurring internally in each country, as changes in macroeconomic policies, than the overall external events. Originality/value It is expected that this study has contributed to a better understanding of the volatility of interest rates and the main factors affecting this market.

Directional dependence via Gaussian copula beta regression model with asymmetric GARCH marginals

ABSTRACTThis article proposes a new directional dependence by using the Gaussian copula beta regression model. In particular, we consider an asymmetric Generalized AutoRegressive Conditional Heteroscedasticity (GARCH) model for the marginal distribution of standardized residuals to make data exhibiting conditionally heteroscedasticity to white noise process. With the simulated data generated by an asymmetric bivariate copula, we verify our proposed directional dependence method. For the multivariate direction dependence by using the Gaussian copula beta regression model, we employ a three-dimensional archemedian copula to generate trivariate data and then show the directional dependence for one random variable given two other random variables. With West Texas Intermediate Daily Price (WTI) and the Standard & Poor’s 500 (S&P 500), our proposed directional dependence by the Gaussian copula beta regression model reveals that the directional dependence from WTI to S&P 500 is greater than that from S&P 500 to WTI. To validate our empirical result, the Granger causality test is conducted, confirming the same result produced by our method.

How do financial features affect volatility forecasts? Evidence from the oil market and other markets

Abstract This study uses six types of generalized autoregressive conditional heteroscedasticity (GARCH) models to estimate the volatility of 28 assets dispersed in the oil, metal, stock, and exchange rate markets, and it explores whether the three financial features of price level, distribution, and leverage effects exist in these four markets. Through an accuracy evaluation, this study also investigates how the financial features affect volatility forecasts and which feature plays the most substantial role in volatility forecasts in each market. Empirical results show that the assets in the oil (resp. exchange rate) market have the greatest (resp. smallest) risk. Moreover , the fat-tailed effect most significantly exists in these four markets, followed by the price level, skewness, and leverage effects. Notably , a negative (resp. positive) volatility elasticity exists in the oil, exchange rate, and stock (resp. metal) markets. Furthermore , both the price level and distribution effects significantly affect the volatility forecasts in the oil market, whereas only the leverage effect slightly affects the volatility forecasts in the metal market. Conversely, the price level, distribution, and leverage effects slightly affect the volatility forecasts in the stock market, whereas no effect can affect the volatility forecasts in the exchange rate market. The price level effect is the most crucial in volatility forecasts in the oil market, whereas the leverage effect is the most crucial in volatility forecasts in the metal and stock markets. Additionally, the GJR-GARCH-N has the best performance in volatility forecasts among the three asymmetric GARCH models.

A Forecast Comparison of GARCH Models and Implied Volatility

This research paper empirically investigates the performance of various models of volatility forecasting. Along with “unconditional standard deviation”, most commonly use measure of volatility, present research considered six others models of volatility forecasting, namely Standard GARCH (Generalized Autoregressive Conditional Heteroscedasticity) model, GJR-GARCH model, Exponential GARCH model (eGARCH), Asymmetric Power GARCH model (apGARCH), Component Standard GARCH model (csGARCH), and Option Implied Volatility model. To forecast volatility, the daily closing values of NSE Nifty 50 (Index) are considered from period 1st April 2002 to 31st March 2015, a span of 13 years. The present study found no single model as the best performing model. Overall performance found GJR-GARCH and option implied volatility having consistence performance. Interestingly, most commonly used model i.e. unconditional standard deviation found to be the worst model among all models studies in present research.

Estimating Stock Market Volatility with Asymmetric Arch, GARCH and Expanded GARCH Models

Purpose: This paper examines the volatility of stock return in Dhaka stock exchange, Bangladesh Methodology: Using Random Walk model (RW), Autoregressive model (AR), Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model, and extensive GARCH model with Normal, and Student t-distribution. The data span for the study from 04 January 2004 to 14 September 2015. Findings: The study revealed that daily return of DSE are leptokurtic, significant skewness, deviation from normality and the return series are volatility clustering. It is also obvious for study that GARCH family can be used to predict volatility of stock return in Dhaka stock exchange (DSE).

Examining Volatility Persistence and News Asymmetry in Soybeans Futures Returns

Four alternative generalized autoregressive conditional heteroscedasticity (GARCH), and three asymmetric GARCH models (EGARCH, TGARCH and APARCH) are used to examine the presence of volatility persistence and news asymmetry in soybeans futures data. Presence of fat tails in the data series resulted in applying Student’s-t and generalized error distributions in addition to Gaussian normal distribution. The results reveal that soybean return series exhibit volatility characteristics typical of a financial time series. The findings of this study indicate that the leverage effect was absent for soybeans suggesting that positive news causes more volatility to the commodity than negative news. Results further suggest that the fit of the GARCH models is improved by applying t-distribution errors. The diagnostic tests reveal that GARCH models are correctly specified and among all the competing models, APARCH (1,3) model with t-distribution performed best in capturing the volatility.

Size and Power of Diagnostic Tests for Asymmetric Garch-Type Models

Generalized autoregressive conditional heteroscedasticity (GARCH)-type models have been successively used to capture the conditional volatility of macroeconomic and financial time series in the past two decades. However, few diagnostic tests are specifically devised to check the adequacy of symmetric multivariate GARCH specifications. Moreover, most practitioners resort to the popular Ljung-Box test indiscriminately, even though the appropriateness of such a test is questionable. In this paper, we investigate the empirical size and power of four diagnostic tests: the Ling-Li test, Ljung-Box test, the Box-Pierce test modified by Tse and Tsui, and the runs test, respectively. We use Monte Carlo simulation experiments over a wide combination of data generating processes and estimation models of bivariate GARCH-type asymmetric models. In the absence of analytically derived diagnostic tests, our simulation results could serve as guidelines for empirical researchers and practitioners in selecting the appropriate diagnostic tests for multivariate asymmetric GARCH models.

Modelling petroleum future price volatility: analysing asymmetry and persistency of shocks

AbstractUnderstanding the nature of volatility in commodity prices warrants adequate attention because such volatility is likely to lead to increased production, search and opportunity costs, as well as accelerate uncertainty and risk, contributing to a slowdown of economic activities. This study examines the asymmetry and persistency in the volatility of a set of petroleum future price returns—namely crude oil, heating oil, gasoline, natural gas and propane—within the framework of a set of non‐linear generalized autoregressive conditional heteroscedasticity (GARCH)‐type models. Specifically, we employ threshold GARCH, exponential GARCH, asymmetric power ARCH and component GARCH models using daily data over the period 1995–2010. The study reveals the following: over the full sample period of 1995–2010, all future price returns show persistent and asymmetric effects of shocks to the volatility but the level of persistency and degree of asymmetry differ product to product; over the subsample 1995–2001, persistency and asymmetry are evident for all series with the exception of gasoline future price returns; the recent subsample of 2002–2010 shows mixed evidence and all series show persistent effects of shocks to the volatility while asymmetry is supported in crude oil and propane only; the study also concludes that based on forecasting performance, no single model can be recommended but different models should be used based on the time periods involved and the nature of petroleum products. These findings also imply that in the presence of asymmetric and persistent volatility, policy measures should be taken to accommodate long lasting effects of shocks to the volatility. And since negative effects of shocks are not fully compensated by positive shocks, counter‐cyclical policies should be taken to counter the pessimistic and optimistic overreactions of businesses to ensure a stable business environment.

Análise da Efetividade de Políticas de Hedge no Mercado de Dólar Futuro no Brasil

Nos últimos anos, o mercado de dólar comercial experimentou grande volatilidade no Brasil. Os episódios de alta forte e rápida do dólar comercial provocaram prejuízos para diversas empresas brasileiras com dívida nesta moeda. A queda do dólar entre 2002 e 2008 provocou perdas para o setor exportador. Nesse contexto, o objetivo deste estudo foi examinar a efetividade do Hedge no mercado futuro de dólar comercial, negociado na BM&FBovespa, no período de dezembro de 2001 a fevereiro de 2009. A determinação da razão de Hedge foi feita de quatro maneiras alternativas: a) ingênua ou 1 – 1, na qual o Hedger toma uma posição totalmente inversa à sua posição à vista; b) MQO – Mínimos Quadrados Ordinários; c) GARCH (Generalized Autoregressive Conditional Heteroscedasticity) Simétrico Bivariado; d) GARCH Assimétrico Bivariado. Os resultados do estudo mostraram que houve uma melhoria significante na efetividade do hedge utilizando modelos GARCH em relação às estratégias ingênuas e MQO

GARCH Models and the Financial Crisis-A Study of the Malaysian Stock Market

Financial market volatility is an important aspect when setting up strategies related to portfolio management, options pricing and market regulation. Occurrence of the global financial crisis of 2007/2008 affected all financial markets around the world and a major concern was about the volatility changes in stock markets. This study has investigated the change in volatility of the Malaysian stock market, with respect to the global financial crisis of 2007/2008, using both symmetric and asymmetric Generalized Autoregressive conditional heteroscedasticity (GARCH) models. Using the Kuala Lumpur Composite Index (KLCI), two periods are selected. The first period is from June 2000, after the recovery of the East Asian crisis, to the end of 2007 and excludes the global financial crisis 2007/2008 and the second period includes the crisis, i.e., from June 2000 to March 2010. AR (4) is found to be the best in modelling the conditional mean and GARCH (1, 1), EGARCH (1, 1), GJR-GARCH (1, 1) for conditional variance. As expected from financial time series, for both periods, the KLCI exhibits stylized characteristics such as leptokurtosis, clustering effect and asymmetric and leverage effect. It is also found that there was a significant increase in volatility and leverage effect but just a small drop in persistency due to the financial crisis.

Impact study of volatility modelling of Bangladesh stock index using non-normal density

This article examines a wide variety of popular volatility models for stock index return, including the random walk (RW), autoregressive, generalized autoregressive conditional heteroscedasticity (GARCH), and asymmetric GARCH models with normal and non-normal (Student's t and generalized error) distributional assumption. Fitting these models to the Chittagong stock index return data from the period 2 January 1999 to 29 December 2005, we found that the asymmetric GARCH/GARCH model fits better under the assumption of non-normal distribution than under normal distribution. Non-parametric specification tests show that the RW-GARCH, RW-TGARCH, RW-EGARCH, and RW-APARCH models under the Student's t-distributional assumption are significant at the 5% level. Finally, the study suggests that these four models are suitable for the Chittagong Stock Exchange of Bangladesh. We believe that this study would be of great benefit to investors and policy makers at home and abroad.