Volatility Series Research Articles

Modelling Volatility of Size Indices Through Garch and Tgarch Models: Evidence from India

Purpose: The objective of this study was to select the best model for modelling the volatility of log return series of four size indices taken from the Indian stock markets. Theoretical framework: Stock markets throughout the world have been designing various kinds of indices depending upon the requirements of the investors. Accordingly, BSE launched some indices, categorized as size indices, depending upon the market capitalization of the stocks listed on their exchange. Such categorization helps the investors in appropriate decision making. Research on these specific indices is scarce and therefore the aim of the present study. Design/methodology/approach: We analyzed the log return series of four indices, namely S&P BSE Sensex, S&P BSE Large cap index, S&P BSE Mid cap Select index and the S&P BSE Small cap index, by fitting the GARCH (1,1) and the TGARCH (1,1) models for the period from 1 Januray 2018 to 31 December 2022. Findings: The research on modelling volatility is voluminous with varied results as per the various assets and datasets utilized for the purpose. Present study confirms the results from previous study that to model the volatility of size indices a GARCH (1,1) model should suffice and there is no betterment achieved by implementing a TAGARCH (1,1) model. Research, Practical & Social implications: We suggest the implementation of a simple GARCH (1,1) model for all those investors who want to invest in Indian size indices in order to model volatility of these indices. Originality/value: Present study is one of its kinds, investigating size indices specifically, through the two of the most popular GARCH-family models in the Indian context.

RETRACTED ARTICLE: Analysis of the impact of the trade war between China and America on the RMB exchange rate under the R-vine copula model from the perspective of the global value chain

AbstractChanges in the RMB exchange rate (ExRate) impact major strategic directions, such as China’s rise, the “Belt and Road” initiative, and RMB internationalization. Studying the impact of the trade war between China and America on the RMB ExRate is significant. This paper constructs an R-vine copula model and studies the influencing factors of the ExRate and the mechanism of the trade war between China and America on the RMB ExRate. The impact of the Trade war between China and America on the RMB ExRate is analyzed in detail. Besides, the R-vine copula model is used to analyze the fluctuations of the ExRate returns of the Association of Southeast Asian Nations (ASEAN). The R-vine copula model is employed to study the volatility of the ExRate return series of ASEAN countries, and the mean of the ExRate dependence coefficients of ASEAN and the estimation results of R-vine copula parameters are analyzed. Finally, the results are obtained. There is a relation between the RMB ExRate and Chinese tariffs in analyzing the impact of the trade war between China and America on the RMB ExRate. There is a negative correlation between tariffs and the RMB ExRate. There is a positive correlation between the value of the RMB and China’s trade surplus without considering the monetary policy. The value of the Chinese currency has risen as China’s trade surplus has grown. Stable foreign exchange reserves can stabilize the RMB ExRate. It is concluded that the economic and trade cooperation between China and ASEAN has become increasingly frequent under the influence of the trade war between China and America through the R-vine copula model analysis. The ExRate risk further increases and the ExRate yield sequence fluctuates significantly. The lower tail dependency coefficient for market tail dependencies is slightly larger than the upper tail dependency coefficient. The impact of bad news on the currency ExRate market is clear. In the context of the current trade war between China and America intensifying, the sharp fluctuations in the RMB ExRate caused by the trade war between China and America have promoted the transformation of economic growth. The shift from commodity export to capital export is realized. The reasons for RMB ExRate changes need to be explored. It is recommended to continuously deepen the reform of the RMB ExRate system and maintain the stable development of the ExRate.

An Efficient GAN-Based Multi-classification Approach for Financial Time Series Volatility Trend Prediction

Deep learning has achieved tremendous success in various applications owing to its robust feature representations of complex high-dimensional nonlinear data. Financial time-series prediction is no exception. Hence, the volatility trend prediction in financial time series (FTS) has been an active topic for several decades. Inspired by generative adversarial networks (GAN), which have been studied extensively in image processing and have achieved excellent results, we present the ordinal regression GAN for financial volatility trends (ORGAN-FVT) method for the end-to-end multi-classification task of FTS. An improved generative model based on convolutional long short-term memory (ConvLSTM) and multilayer perceptron (MLP) is proposed to capture temporal features effectively and mine the data distribution of volatility trends (short, neutral, and long) from given FTS data. Meanwhile, ordinal regression is leveraged for the discriminator to improve the multi-classification performance, making the model more practical. Finally, we empirically compare ORGAN-FVT with several state-of-the-art approaches on three real-world stock datasets: MICROSOFT(MSFT), Tesla(TSLA), and The People’s Insurance Company of China(PAICC). ORGAN-FVT demonstrated significantly better AUC and F1 scores, at most 20.81% higher than its competitors.

Improving financial volatility nowcasts*

ABSTRACT Recently, a simple nowcasting model for volatility has been proposed by Breitung and Hafner [2016. A Simple Model for Now-Casting Volatility Series. International Journal of Forecasting 32 (4): 1247–1255. doi:10.1016/j.ijforecast.2016.04.007]. They suggest a model in which today's volatility is not only driven by past returns, but also by the current information from the same day. Empirical results demonstrate the relevance of the current squared return for volatility nowcasting. Their model obeys an ARMA representation estimable by maximum likelihood. However, their estimation approach builds on a number of simplifications and we suggest improvements. Rather than assuming normality of the innovations in the ARMA representation for highly skewed and leptokurtic log-squared returns, we take non-normality explicitly into account. Contrary to most situations regarding volatility estimation and forecasting, the distribution actually plays a crucial role in the construction of volatility nowcasts. We devise an exact maximum likelihood estimator which offers significant improvements in estimation efficiency and volatility nowcast accuracy in finite samples. In our empirical application, we investigate five major international stock markets from 2000 to 2019 (including sub-samples relating to the Great Financial Crisis). The results suggest that our estimation approach significantly outperforms the one by Breitung and Hafner [2016. A Simple Model for Now-Casting Volatility Series. International Journal of Forecasting 32 (4): 1247–1255. doi:10.1016/j.ijforecast.2016.04.007] in all cases. Financial volatility can be nowcasted more accurately by applying our suggested estimation approach.

LSTM-GARCH Hybrid Model for the Prediction of Volatility in Cryptocurrency Portfolios.

In the present work, the volatility of the leading cryptocurrencies is predicted through generalised autoregressive conditional heteroskedasticity (GARCH) models, multilayer perceptron(MLP), long short-term memory (LSTM), and hybrid models of the type LSTM and GARCH, where parameters of the GARCH family are included as features of LSTM models. The study period covered the scenario of the World Health Organization pandemic declaration around March 2020 at hourly frequency. We have found that the different variants of deep neural network models outperform those of the GARCH family in the sense of the hetorerocedastic error, and absolute and squared error(HSE). Under the sharpe ratio, the volatility forecasting of a uniform portfolio at long horizons systematically outperforms the stablecoin Tether, which is considered here as the risk-free asset. Also, including transaction volume helps reduce the value at risk or loss probability for the uniform portfolio. Moreover, in a minimum variance portfolio, it is observed that before the pandemic declaration, a large proportion of the capital was allocated to bitcoin(BTC). In contrast, after March 2020, the portfolio is more diversified with short positions for BTC. Moreover, the MLP models give the best predictive results, although not statistically different in accuracy compared to the LSTM and LSTM-GARCH versions under the Diebold-Mariano test. In sum, MLP models outperform most stylised financial models and are less computationally expensive than more complex neural networks. Therefore, simple learning models are suggested in highly non-linear time series volatility forecasts as it is the cryptocurrency market.

Estimation of stock market index volatility using the GARCH model: Causality between stock indices

This paper aims to model the volatility of returns for selected stock indices and examine the causal relationships between the markets using the historical daily prices of the Financial Times Stock Exchange (FTSE), Bursa Malaysia Kuala Lumpur Composite Index (KLCI), the Indonesia Stock Exchange Index (LQ45), and the Stock Exchange of Thailand (SET) from January 2008 to November 2019. The study employs univariate GARCH models that are prominent in capturing the volatility clustering of financial instruments in association with the Box–Jenkins methodology for better estimation. Generally, the ARMA-GARCH model is used to capture the volatility series, while the Granger causality test examines the causal directions between the markets. The findings revealed leverage effects on the markets, with the outperformance of the EGARCH in analyzing the empirical properties of stock returns. An initial test that yielded positive correlations suggests the existence of co-movement between the derived volatility series. The study concluded bidirectional causal relationships between the selected markets, and based on the resulting relationships, it is proposed that supervision of markets among the ASEAN members could be advantageous in predicting the corresponding market performance.

Research on adaptive control technology of wind power generation based on AI data mining

This paper introduces the current achievement and control technology of wind power generation industry. By analysing the characteristics and influencing factors of wind power generation, we proposed an ultra-short-term wind power forecasting network structure which can work well in wind power forecasting. Based on the characteristics of the datasets, a variety of mainstream neural network structures are compared, our proposed method is suitable for high volatility series is further developed. The structure integrates CNN, RNN and attention mechanism, which alleviates the problem of prediction curve lag to a certain extent and effectively avoids virtually useless prediction.

Analysis of Price Behavior in Sri Lankan Vegetable Market

Vegetables are important source of nutrient for the Sri Lankan population and both farmers and consumers are adversely affected by vegetable price volatility. The lack of price analysis and forecasting has made it difficult to establish an effective early warning system for the vegetable farming sector in Sri Lanka. This study investigates the price behaviour of selected fresh vegetables - carrot, cabbage, and tomato - and forecasts the future prices and volatilities using time series techniques. Analysis of weekly price data from 1997 to 2018 revealed that all three - price series had one structural break, but none coincided with the policy change when the government introduced fertilizer subsidies for vegetable producers in the agriculture sector. The autoregressive integrated moving average (ARIMA) model estimations show that the best model for forecasting carrot price is ARIMA (3,1,2) (0,0,2)[52]* capable of predicting retail prices at 71% accuracy while the best model for cabbage prices is ARIMA(1,1,1)(0,0,1)[52] with a prediction accuracy of 55%. All three-price series exhibit serial correlation in residuals; hence GARCH estimations were used to model and predict volatility. Of the fitted ARMA GARCH models, the best model for estimating the volatility of carrot and cabbage were GARCH (1, 2) ARMA (3, 2) and GARCH (1,1) ARMA (3 ,2), respectively. The volatility predictions for the first ten weeks for the year 2019 indicate a gradual decrease in volatility in the carrot price series whilst a gradual increase in volatility in the cabbage price series.

Investigating the interrelation between sovereign CS and financial series using wavelet coherence analysis: case of Turkey

Purpose- Financial agents pay a great attention to the relationship between credit default swaps (CDS) and other financial assets. Being considered as a financial asset and then as a possible component of a portfolio, CDS’s relationship with other financial assets is crucial in order to compose an efficient portfolio which should contain the least correlated assets. Our study, provide a different approach than the bulk of existing literature which consider the relation between returns/levels or between volatilities and their impact on the volatility (leverage effects) as well as volatilities may affect returns/levels/rates in BIST100 by focusing on several economic and financial determinants for the investment horizon of investors and policymakers for their decision-making processes. Methodology- We investigate the relationship between sovereign CDS spread (level and volatility) and (return/level and volatility) of financial series (bond yield, currency exchange, stock index) by applying wavelet coherence analysis to capture the interrelation between economic and financial determinants during different investment horizons due to the account of the time-frequency domain. Findings- According to empirical results, firstly, the direction of causality between series depends on the frequency domain. A series may cause the other series at medium term and may be influenced by this second series at short term. Secondly, the relationship between the CDS spread (volatility) and retained financial series changes through time while depending the type of financial asset. Lastly, the interconnectedness between returns/rates/levels is different from the relationship between volatilities and between returns/rates/levels and volatilities. Conclusion- Our findings show that the CDS 2-year spread and interest rates have a positive relationship most of the time in the medium term, mostly in the 2017-2019 period, while CDS spreads and the USDTRY have similar impacts in the medium and early long term. Despite having previously observed such a positive relationship between CDS and other economic determinants, CDS spreads have a negative impact on stock exchange returns in the medium term, spanning nearly a decade from 2008 to 2016. Keywords: Time series models, wavelet coherence analysis, sovereign CDS, foreign exchange, bond interest rate, stock market returns JEL Codes: C22, F31, G12

A positive/negative average real variability index and techno-economic analysis of a hybrid energy storage system

The energy storage systems are considered the prime candidates to increase the stability and penetration of renewable energy sources. However, their cost and reliability may prove to be significant obstacles for their development. A solution that is gaining momentum and can augment performance and reduce costs is hybrid energy storage systems (H-ESS). In the current study, two novel indexes are introduced aiming to capture the connection between the net demand time series volatility and a H-ESS with fast response characteristics. A new methodological framework to derive the nominal values of a H-ESS coupled to a PV/Wind system is presented. The proposed H-ESS comprises a lithium-ion battery and superconducting magnetic energy storage (SMES). The flywheel energy storage (FES) is also considered instead of the SMES to compare the performance and economic viability with the fast reserve incentive of a mature and developing technology. The techno-economic viability of the system with/without H-ESS is calculated by the levelized cost of electricity (LCOE), revealing that with a fast reserve incentive, the integration of SMES at an LCOE of 36.734 €/MWh is the more favorable configuration compared to FES at 36.756 €/MWh and only battery at 40.367 €/MWh. A sensitivity analysis for the SMES and FES is conducted, concluding that we can achieve 0.84 % and 0.87 % reduction in the LCOE of SMES and FES, respectively, if the costs are halved. Furthermore, the two novel indexes verify the intuition that a net demand time series with high volatility favors a H-ESS with fast response characteristics, which can be quantified to an LCOE of 62.791 €/MWh and 64.712 €/MWh without fast reserve for SMES/battery and only battery configurations, respectively. This work can be utilized by market participants and/or potential investors to optimally determine their portfolios.

Volatile time series forecasting on the example of the dynamics of the Dow Jones index

The paper discusses a new predictive model of a fuzzy volatile time series, in the framework of which a new approach to the data fuzzification is proposed as the results of observations based on “Sоft Measurements”. As an example, the index of the Dow Jones Industrial Average is chosen, the readings of which are established by usual arithmetic averaging of cоntextual indicators. This allows to consider the daily readings of the Dow Jones index as weakly structured, and to interpret the dynamics of its change as a fuzzy time series. The data fuzzification is realized by applying the fuzzy inference system that provides the values of the membership functions of the appropriate fuzzy sets on the universe covering the set of Dow Jones index for the period from June 15, 2018 to October 10, 2019. The prоpоsed predictive mоdel is based on the identified internal relatiоnships, designed as 1st оrder fuzzy relatiоns between evaluation criteria (or fuzzy sets) that describe weakly structured Dow Jones indexes. At the end of the study, the proposed model is evaluated for adequacy using the statistical criteria MAPE, MPE and MSE.

Testing for Short-run and Long-Run Asymmetric Causality for two Volatile Series: An Extension of Frequency Domain Approach

Testing for Short-run and Long-Run Asymmetric Causality for two Volatile Series: An Extension of Frequency Domain Approach

Сравнение моделей прогноза волатильности криптовалют и фондового рынка

The article compares GARСH and HAR models for 1 day ahead forecasting performance of the realized volatility of financial series. As an example, the cryptocurrency with the largest capitalization, Bitcoin, was chosen. Its realized volatility is calculated from intraday (24 hours) data, using the closing values of five-minute trading intervals. The paper proposes a method for calculating realized volatility for the case of gaps in 5-minute intraday data. This makes it possible to achieve comparability of the daily values of the realized volatility of assets with different trading times. All days of the week are almost equally present among the days selected for forecasting. For comparison, a stock market asset was chosen, E-mini S&P 500, a futures contract that is traded 23 hours a day. We use data from 01/01/2018 to 12/29/2021. Since there could be (and were) structural changes in the markets in this interval, the models are evalua­ted in rolling windows 399 days long. For each series 810 GARCH models and 46312 HAR models are compared. The MCS test is used to select the best models (at the significance level of 0,01). It is shown that GARCH models are inferior to HAR models in the accuracy of forecasting both the realized volatility of Bitcoin and the E-mini S&P 500. At the same time, the relative accuracy of the forecast of the realized volatility of Bitcoin is higher than the accuracy of the forecast of the realized volatility of the E-mini S&P 500 futures. The smallest relative errors for Bitcoin and E-mini S&P 500 realized volatility forecasts are 29,51% and 36,12%, respectively.

PV power prediction based on AO-VMD-RF-Informer

Due to the strong volatility of PV power, PV grid-connected may have an impact on the safe and stable operation of the power system, so accurate prediction of PV power is of great significance to the operation and maintenance of the power system. In order to improve the prediction accuracy of photovoltaic power, an ultra-short-term photovoltaic power prediction method was studied by combining the Aquila Optimizer (AO) algorithm, the Variational Mode Decomposition (VMD), the Random Forest (RF) and the Informer prediction model. Firstly, the VMD parameters are optimized by AO to reduce the adverse effects of human-set parameters on the prediction accuracy; the optimized VMD is used to decompose the original PV power series into multiple sub-sequences to reduce the volatility and complexity of the original power series; then, the RF feature selection method is used to screen out the meteorological features of strong relevance for each sub-sequence to further reduce the feature dimensions and the model runtime and ensure the effectiveness of the input features. Finally, the Informer model is used to deeply mine the potential time series features of each subsequence for prediction, and the predicted values of each subsequence are superimposed and reconstructed to obtain the final prediction results. The simulation results show that the method in this paper has high prediction accuracy, and compared with the original Informer, the MAE is reduced by 49.14% and the RMSE is reduced by 47.64%.

Asymmetric volatility spillover between oil prices and regional renewable energy stock markets: A time-varying parameter vector autoregressive-based connectedness approach

<abstract> <p>The rapid expansion of renewable energy sources and their integration into the energy mix has generated scholarly interest in comprehending the interplay between renewable and conventional energy markets. This research aims to examine the (a)symmetric volatility spillover between the oil market and various regional renewable energy stock markets, namely the US, Europe and Asia. To achieve this objective, we employ the time-varying parameter vector autoregressive-based connectedness (TVP-VAR) approach, which allows analysing the interconnection and transmission of shocks between the different markets. Based on an analysis of daily data relative to the different regional renewable energy stock markets and international oil prices, the findings suggest the presence of a dynamic volatility connectedness between the green and brown energy stock markets. The extent of connectedness is contingent upon the specific regional renewable energy market under consideration. Moreover, the decomposition of the volatility series into <italic>good</italic> and <italic>bad</italic> volatility emphasizes an asymmetric pattern, which becomes more pronounced during periods of major events. On average, the oil market and the Asian renewable energy stock market are net receivers of volatility shocks. In contrast, the US and European renewable energy stock markets are net transmitters of shocks. Our findings provide investors with valuable insights for portfolio design and risk management decisions.</p> </abstract>

A Long Memory of Financial Information in Korean Stock and Futures Markets

This paper aims to examine the long memory behavior of financial information (i.e., returns and volatilities) of the spot and futures contracts of the KOSPI 200 index. To this end, we use the novel technique of the wavelet OLS estimation devised by Jensen (1999). For the financial information of return series, we show no significant long memory pattern. Meanwhile, for the information of the volatility series, we find a clear pattern of a long memory regardless of applied wavelet filters.

EXAMINING MEAN AND VOLATILITY SPILLOVER IN FOREIGN EXCHANGE MARKETS: THE TURKISH CASE

Bu çalışmada Euro, İngiliz sterlini, Japon yeni, Brezilya reali, G.Afrika randı, Meksika pesosu ve G.Kore wonu ile Türk lirası arasındaki getiri ve volatilite yayılımı Cheung-Ng (1996) testi ile incelenmiştir. Analizlerde volatilite serilerindeki yapısal kırılmalar da dikkate alınmıştır. Volatilitenin modellenmesinde FIGARCH modelinden, volatilite serilerindeki yapısal kırılmaların belirlenmesinde ise Bai ve Perron (1998, 2003) testinden yararlanılmıştır. Getiri yayılımına ilişkin bulgular; Japon yeni, İngiliz sterlini, Meksika pesosu ile G.Afrika randından Türk Lirasına doğru tek yönlü bir getiri yayılımının söz konusu olduğunu; Türk lirası ile Brezilya reali, Euro ve G.Kore wonu arasında ise çift yönlü bir getiri yayılımının bulunduğunu göstermektedir. Volatilite yayılımına ilişkin bulgular ise inceleme kapsamındaki tüm para birimlerinin volatilitesinin Türk lirası volatilitesi üzerinde etkili olduğunu, fakat Türk lirası volatilitesinin daha çok kendi iç dinamiklerinden kaynaklandığını göstermektedir. Çalışma bulguları hem uluslararası yatırımcılar hem de politika yapıcılar için önemli sonuçlar içermektedir.

INFORMATIONAL EFFICIENCY OF THE US MARKETS FOR IMPLIED VOLATILITY BEFORE AND AFTER THE COVID-19 PANDEMIC

The objective of this work is to assess informational efficiency in four US markets for implied volatility. This has been pursued using daily data over 2015 to 2021 and a composite index that accounts for three possible sources of inefficiency associated with long-range dependence, short-range dependence, and entropy. The dominant pattern of long-range dependence has been that of anti-persistence both before and during the pandemic. The same applies for short-range dependence, especially before the pandemic. The presence of anti-persistence is an indication of investors’ over-reaction to incoming information and implies that oscillatory trading strategies have been probably more successful that trend-following ones. During the Covid-19 pandemic, the entropy decreased in all cases suggesting that the four implied volatility series became more predictable; the intensity, however, of long-range and short-range dependence remained largely unaffected. As a result of these developments, the informational efficiency in at least two markets (those related to stock and to crude oil) fell.

The chaotic, self-similar and hierarchical patterns in Bitcoin and Ethereum price series

Bitcoin (BTC) and Ethereum (ETH), pioneering public blockchains implementations, are two fundamental levers to register and transfer digital value. This article studies the structure of their daily price volatility time series following a multifaceted approach: first, it examines the existence of chaoticity and fractality in the time series. Obtained results confirm that the BTC and ETH price volatility series present signs of chaoticity, persistence of a long-term correlation and multifractality. Second, it analyses the corresponding visibility graphs associated with these time series using complex network theory. The undirected and connected complex networks, spawned by their natural visibility graphs (VGs) and horizontal visibility graphs (HVGs), present a hierarchical structure. These networks, especially the HVGs, confirm the fractality of the originating time series. The study of HVGs also confirms a lack of uncorrelated randomness in the originating BTC and ETH price series. This paper validates the value of visibility graphs as useful proxies to better understand complex time series, in this case, related to public blockchain implementations.

The Impact of Fundamentals on Volatility Measures of Agricultural Substitutes

AbstractThis study builds upon the existing literature on the Working curve and backwardation to explore the impact of storage regimes on the volatility measures of substitute agricultural commodity markets. We investigate the impact of commodity fundamentals (storage regime and stocks-to-use ratio), commodity-specific financial variables (options hedging pressure-long and -short), world economic activity, market-wide volatility index, seasonality, and time-to-maturity on nearby and deferred implied volatility (IV) series of selected commodity pairs of corn-soybean and winter wheat-spring wheat. Our work confirms that, in some cases, grain and oilseed IV derived from options premia respond to shocks in commodity (and substitute commodity) fundamentals which are in line with the behaviour of volatility in futures markets. Own-storage regime effects on price variability are stronger in the selected markets, while spillover effects from substitute commodity storage regimes show a modest impact on volatilities. We also find some evidence for the stocks-to-use ratio of both corn and soybean to impact both their own and each other’s IV, while options hedging pressure has some impact only on wheat IVs.