Autoregressive Conditional Heteroskedasticity Model Research Articles

Forecasting the Volatility of Real Residential Property Prices in Malaysia: A Comparison of Garch Models

Abstract The presence of volatility in residential property market prices helps investors generate substantial profit while also causing fear among investors since high volatility implies a high return with a high risk. In a financial time series, volatility refers to the degree to which the residential property market price increases or decreases during a particular period. The present study aims to forecast the volatility returns of real residential property prices (RRPP) in Malaysia using three different families of generalized autoregressive conditional heteroskedasticity (GARCH) models. The study compared the standard GARCH, EGARCH, and GJR-GARCH models to determine which model offers a better volatility forecasting ability. The results revealed that the GJR-GARCH (1,1) model is the most suitable to forecast the volatility of the Malaysian RRPP index based on the goodness-of-fit metric. Finally, the volatility forecast using the rolling window shows that the volatility of the quarterly index decreased in the third quarter (Q3) of 2021 and stabilized at the beginning of the first quarter (Q1) of 2023. Therefore, the best time to start investing in the purchase of real residential property in Malaysia would be the first quarter of 2023. The findings of this study can help Malaysian policymakers, developers, and investors understand the high and low volatility periods in the prices of residential properties to make better investment decisions.

Exploring Volatility and Spillover Effects between African Sovereign Bond Markets and Global Long-Term Interest Rates

The study investigated the existence of volatility and spillover effects between sovereign bond returns of South Africa and Ethiopia and the world’s long-term interest rate using multivariate generalized autoregressive conditional heteroskedasticity model. The results showed that volatility from the long-term world interest rate negatively affects the Ethiopian sovereign bond market. The results also showed a one-way spillover from South Africa's market to the U.S. long-term market, then from the U.S. to Ethiopia's market, and further from Ethiopia's to South Africa's market. However, no bidirectional spillover was observed within these markets. Besides, both African markets display high volatility persistence. Besides, the markets have a weak or insignificant correlation with the world’s long-term interest rate. Volatility in the markets is significantly affected by their respective past shocks or volatilities. Finally, it has forwarded policy inputs that should be tailored to the specific economic and financial context of each country.

Comparison Of Artificial Neural Network Backpropagation and Garch Methods In Predicting Stock Price (Case Study: Indosat Shares 2012 – 2022)

Abstract: One of the problems encountered in the forecasting process is the problem of heteroscedasticity. Heteroscedasticity occurs a lot, especially in stock data. Pt Share Price Indosat (tbk) from March 6 2012 – January 18 2022 has fluctuated from time to time, so the variance is heteroscedasticity. The Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model and Artificial Neural Network Backpropagation (ANNBP) are methods that can be used on data with heteroscedasticity. The aim of this research is to obtain models and forecasting results from GARCH and ANN Backpropagation. In this study, the two models were compared based on the smallest MAPE value. This study uses daily data on the closing of Indosat shares. Forecasting is done on Indosat stock closing data, the total data is 2453 data divided into two parts, namely 80% training data totaling 1962 data and 20% training data totaling 491 data. Forecasting results from the GARCH model obtained a MAPE value of 11.04%, and the ANN Backpropagation model with 7 input layers, 20 hidden layers, obtained a MAPE value of 7.01%. Thus, the best model for predicting Indosat's share price in this study is the backpropagation model.

The ripple effects of CBDC-related news on Bitcoin returns: Insights from the DCC-GARCH model

Central bank digital currencies (CBDCs) have emerged as a potential substitute for current payment methods, and, as such, major announcements, events and policy discussions regarding CBDCs have the potential to influence cryptocurrency returns. In light of this, the present study undertakes an in-depth analysis of the CoinMarketCap data between August 1, 2017 and April 1, 2022 by implementing the dynamic conditional correlation-generalized autoregressive conditional heteroskedasticity (DCC-GARCH) model. The study reveals a noteworthy influence of news and events related to CBDCs on Bitcoin returns. Precisely, CBDC uncertainty index and CBDC attention index have resulted in significant fluctuations in Bitcoin returns, indicating that positive news can result in significant Bitcoin returns. The findings suggest that future expectations of investors regarding cryptocurrencies are shaped by CBDC-related news and events.

Discussing energy volatility and policy in the aftermath of the Russia–Ukraine conflict

The ongoing Russo–Ukrainian War has highly affected energy markets in the EU and worldwide, with different EU- and country-level emergency policy measures being advanced to tackle high energy prices. Despite the progress in green energy initiatives and the race toward climate neutrality by 2050, high energy prices are a matter of concern for all EU countries in the short-to-medium term. The current study investigates the energy price volatility in the aftermath of the Russia–Ukraine conflict, which is of high interest for designing effective government measures (such as monetary and energy policies) addressing the consequent changes occurring in employment, economic activity, commodity and food prices, and, ultimately, sustainable development. For the empirical analysis, we employed generalized autoregressive conditional heteroskedasticity models to capture the volatility of the following energy commodities: Brent crude oil, TTF natural gas, and UK natural gas. The empirical results reveal an elevated degree of persistence of the volatility, namely, that the generalized autoregressive conditional heteroskedasticity term has a slow decay and pronounced fluctuations for all the energy products. The vulnerability of the EU’s energy policy to geopolitical factors is highlighted, especially for gas, which could be due to its dependence on Russian imports. Moreover, the sanctions imposed by the EU on Russia, namely the sixth package of sanctions, have a minimal immediate effect on stabilizing energy returns. The study offers several policy recommendations to improve the resilience of the EU’s energy sector.

Goodness‐of‐fit tests for the multivariate Student‐t distribution based on i.i.d. data, and for GARCH observations

We consider goodness‐of‐fit tests for the multivariate Student's t‐distribution with i.i.d. data and for the innovation distribution in a generalized autoregressive conditional heteroskedasticity model. The methods are based on the empirical characteristic function and are relatively easy to implement, invariant under linear transformations, and globally consistent. Asymptotic properties of the proposed procedures are investigated, while the finite‐sample properties are illustrated by means of a Monte Carlo study. The procedures are also applied to real data from the financial markets.

Drivers of WACR-Repo Rate Spread: An Empirical Analysis of India

The spread between the policy rate and operating target rate of the Reserve Bank of India (RBI) has consequences on the transmission of monetary policy signals. Using weekly data from 2011 to 2020 (504 observations), the study employs variants of General Auto Regressive Conditional Heteroskedastic model to comprehensively analyse the factors that drive this spread. Besides the factors retrieved from the literature, this study also gauges the impact of COVID-19 on the spread and its volatility. The results indicate that the spread is significantly impacted by its lag, volume of transactions in the counterpart markets, distribution pattern of the liquidity by the central bank, foreign market intervention, liquidity adjustment facility (LAF), currency in circulation and reserve requirement while market expectations and government cash balances are found to be non-significant. Besides, announcement effect of COVID and changes in key rates are found to be significant. The results of Exponential GARCH do not find significant asymmetry in response to positive and negative shocks on the spread. Volatility in the spread is comparatively more responsive to its lags than to the market shocks.

The Estimation of Heavy Tails in Non-linear Models

A generalized student t distribution technique based on estimation of bilinear generalized autoregressive conditional heteroskedasticity (BL-GARCH) model is introduced. The paper investigates from empirical perspective, aspects of the model related to the economic and financial risk management and its impacts on volatility forecasting. The purposive sampling technique was applied to select four banks for the study, namely First Bank of Nigeria (FBN), Guaranty Trust Bank (GTB), United Bank for Africa (UBA) and Zenith Bank (ZEB). The four banks are selected, because their daily stock prices are considered to be more susceptible to volatility than those of other banks within the sampled period (January 2007–May 2022). The data collected were analyzed using MATLAB R2008b Software. The results show that the newly introduced generalized student t distribution is the most general of all the useful distributions applied in the BL-GARCH model parameter estimation. It serves as a general distribution for obtaining empirical characteristics such as volatility clustering, leptokurtosis and leverage effects between returns and conditional variances as well as capturing heavier and lighter tails in high frequency financial time series data.

Currency portfolio behavior in seven major Asian markets

This study explores the returns and volatility patterns of currency portfolios (CP) in seven major Asian markets during noncrisis and crisis periods. The CP is developed by investing or divesting in the top three forward premiums or discount currencies in these markets and is rebalanced quarterly. The main findings are as follow: (1) Stochastic volatility models perform better than generalized autoregressive conditional heteroskedasticity models. (2) Spillover effects from the stock market on CP are observed for the examined periods. (3) The stochastic dominance analysis revealed that CP outperformed most stock markets.

Exchange Rate and Stock Prices Volatility Connectedness and Spillover during Pandemic Induced-Crises: Evidence from BRICS Countries

This paper investigated exchange rate and stock price volatility connectedness and spillover in Brazil, Russia, India, China, and South Africa (BRICS) during pandemic-induced crises. We first extracted volatility using the Generalized Auto-Regressive Conditional Heteroskedasticity (GARCH) model. Then volatility connectedness and spillover were investigated by using (Diebold and Yilmaz, International Journal of Forecasting, 28(1), 57–66, 2012) method. We find that exchange rate volatility and stock return volatilities are connected during pandemic-induced crises. The study also finds volatilities spillover among countries in the sample. Russia has strong volatility connectedness with India in these financial markets. The direction of volatility spillover is from Russia to India. Similarly, Brazil has strong volatility connectedness with South Africa and the direction volatility spillover is from Brazil to South Africa. Finally, China has a weak volatility connection with the remaining BRICS countries. Thus, the volatility transfer in these financial markets and across BRICS countries has economic implications.

Long-term probabilistic streamflow forecast model with “inputs–structure–parameters” hierarchical optimization framework

Accurate, reliable, and stable streamflow forecasts are essential for risk assessment and decision making in water resources management. Owing to the limited value of deterministic forecasting, probabilistic forecasting incorporating probability and confidence intervals is achieved by optimizing the predictors and forecast equation while identifying forecast error features. The combination of deterministic forecasts and error identification can facilitate water resources management. This study improved the traditional forecast model’s inputs selection, structure designation, and error parameters calibration to propose a long-term probabilistic streamflow forecast model with hierarchical optimization of the predictors, forecast equation, and errors characteristics. The model develops information entropy theory to screen the driving predictor set, the long short-term memory (LSTM) model to conduct deterministic forecasts, and the generalized autoregressive conditional heteroskedasticity (GARCH) model to identify time-varying errors. The proposed model, used in some case studies to forecast the monthly streamflow of two lakes, Hongze and Luoma lakes in China, was evaluated using the multidimensional index method considering “accuracy–reliability–stability” performance. The results revealed the following: (1) Predictor screening based on information entropy investigates the statistical characteristics of predictors, thereby improving the reliability of streamflow forecasts. (2) The LSTM model exploits the response between the driving predictors and streamflow, whereas GARCH model identifies the time-varying features of forecast errors effectively, which reduces the probability of forecast failure and increases the accuracy and stability of probabilistic forecasts. (3) Under current meteorological observation conditions and forecasting capability, the proposed forecasts can extend the maximum forecast lead time from 1 month to 3 months. (4) The proposed model improves the accuracy (root mean square error: 6.7%–34.8%), reliability (Brier score: 15.3%–27.9%), and stability (mistaken distance: 36.4–52.6%) in comparison with the benchmark of the two case studies, indicating that “inputs–structure–parameters” hierarchical optimization provides effective forecast information for water resources management.

A Behavioral Interpretation of Volatility Patterns in Brazilian Stock Market: Analysis of Pre and Post-COVID-19 Periods from 2019 to 2021

There has recently been an increasing number of new investors in the Brazilian stock markets, providing new asset allocation opportunities as well as challenges in terms of individual judgment and decision-making. The departure point of this paper is the conjecture that the Ibovespa (Brazilian stock exchange index) might reveal asymmetric volatility patterns during the January 2019 to February 2021 period that call for behavioral financial explanations. More specifically, this paper aims to subject this bold conjecture to an empirical test. It draws on a threshold autoregressive conditional heteroskedasticity (TARCH) model to investigate the existence of asymmetric reactions in the period under study. Our findings suggest that the traditional approach to finance fails to explain volatility patterns in the Brazilian stock market behavior, since they do not fit well with predictions based on the efficient market hypothesis. Inspired by Vasileiou’s (2021) account of what occurred with the New York Stock Exchange during COVID-19, we then embark on a brief survey of the vast literature on applied behavioral economics and finance to provide an interpretation of the heuristics and biases that may be non-negligible mental mechanisms underlying the Ibovespa’s detected asymmetric responses.

Applications of Long-Memory and Structure Breaks for Carbon Indexes

This paper aims to investigatethe long-memory properties of four carbon indexes by utilizing the autoregressive frictionally integrated moving average–fractionally integrated general autoregressive conditional heteroskedasticity models. First, this study discovered a significant long-memory effect for two carbon indexes such as CCX and JOI, whereas others like CER and EUA possess intermediate memory in the returns. Second, the multiple structure breaks in the four carbon indexes were examined using the iterated cumulative sum of squares algorithm. Evidence shows that the sudden shifts are mainly attributed to macroeconomic factors, energy dynamics, and political policies.

Using EGARCH models to predict volatility in unconsolidated financial markets: the case of European carbon allowances

The growing interest and direct impact of carbon trading in the economy have drawn an increasing attention to the evolution of the price of CO2 allowances (European Union Allowances, EUAs) under the European Union Emissions Trading Scheme (EU ETS). As a novel financial market, the dynamic analysis of its volatility is essential for policymakers to assess market efficiency and for investors to carry out an adequate risk management on carbon emission rights. In this research, the main autoregressive conditional heteroskedasticity (ARCH) models were applied to evaluate and analyze the volatility of daily data of the European carbon future prices, focusing on the last finished phase of market operations (phase III, 2013–2020), which is structurally and significantly different from previous phases. Some empirical findings derive from the results obtained. First, the EGARCH (1,1) model exhibits a superior ability to describe the price volatility even using fewer parameters, partly because it allows to collect the sign of the changes produced over time. In this model, the Akaike information criterion (AIC) is lower than ARCH (4) and GARCH (1,1) models, and all its coefficients are significative (p < 0.02). Second, a sustained increase in prices is detected at the end of phase III, which makes it possible to foresee a stabilization path with higher prices for the first years of phase IV. These changes will motivate both companies and individual energy investors to be proactive in making decisions about the risk management on carbon allowances.

The Relationship between the Gold Price, Crude Oil Price, Exchange Rate and Chinese Stock Market Indexes

One of the trickiest problems for investors is how the financial and commodity markets interact with each other. The volatility in one market might affect the price of the other market. This essay aims to clarify the relationship between gold, crude oil, exchange rates, and Chinese stock market indices. In order to do this, the Shanghai Stock Exchange Index and the China Industrial Index, two indices that reflect the Chinese financial market, were subjected to the DCC-GARCH model (Generalized Autoregressive Conditional Heteroskedasticity Model). By capturing the dynamic correlations of the time series' variance, covariance, and correlation coefficient, this model aids in illuminating the long-run dynamic correlation between returns. The findings show that gold and crude oil are positively correlated with both the Shanghai Stock Exchange Index and the China Industrial Index, while the correlations between the USD CNY exchange rate and the two Chinese stock indices are negatively correlated, using monthly data from 2000-01-03 to 2022-08-30. This paper shows that stock prices in China are weakly linked to international commodity markets and exchange rate with the US. Additionally, throughout the observation period, financial crisis (2008) and the COVID-19 pandemic breakout (2020) enhanced comovement between the Chinese stock market, commodities, and exchange rate in general.

Corn price volatility and producer income protection

O Objective: To estimate the risk indicator of the future price of yellow corn #2 at the Chicago Futures Exchange (USA) regarding the spot price of white corn in the main producing regions in Mexico through the financial volatility indicator. Methodology: The research used the returns of the monthly time series corresponding to the spot price of white corn from January 1998 to December 2020, considering five producer-consumer regions of Mexico and the future price of yellow corn #2 as listed on the Chicago Stock Exchange. To quantify volatility, the generalized autoregressive conditional heteroskedasticity model of order (1,1) was estimated. Results: The yellow corn #2 volatility indicator was 0.9870 (future price). In the case of the spot price of white corn in Mexico, the volatility was 0.7977 for the national price, 0.3385 for the central region, 0.3206 for the western region, and 0.0078 for the southeast region. Implications: The high volatility of yellow corn #2 (close to unity) shows that the international market for this commodity is riskier than the national market or regional markets in Mexico. Conclusions: The national white corn market proved to be riskier than the west, center, and southeast regional markets, which have a higher volatility indicator.

Analyzing different types of asymmetric responses in terminal gate prices of petrol and diesel across major Australian seaports: Implications for fuel price monitoring strategies

This paper examines the sign and size-dependent asymmetry in wholesale petrol and diesel prices in the long and short run across seven major distribution centres in Australia using daily data from 1 January 2004 to 18 November 2022. This study adopts a composite threshold mean model that is augmented with an asymmetric power autoregressive conditional heteroskedasticity (APARCH) model or an exponential general autoregressive conditional heteroskedastic (EGARCH) model to capture volatility clustering in high-frequency data. Inter alia, the results reveal that the long-run (adjustment) asymmetry is more widespread for unleaded petrol prices than diesel prices. Melbourne and Perth are the only two outport terminals that do not exhibit short- or long-run asymmetry for both diesel and petrol. In terms of the adjustment asymmetry, terminal gate prices for petrol in Darwin, Hobart, and Sydney are the least competitive, whereas the opposite is true for Adelaide, Brisbane, and Perth.

The asymmetry of shocks in crude oil price in Nigeria

While studies in Nigeria have focused on the consequences of oil price shocks on the economy, they implicitly assumed that oil price changes are symmetrical without specifically testing for asymmetry. This study concentrated on the shock asymmetries and crude oil price volatility in Nigeria. Used were time series data for the years 1981 to 2022 from the Statistical Bulletin of the Central Bank of Nigeria and the World Bank. The data were analyzed using the Exponential Generalized Autoregressive Conditional Heteroskedasticity (EGARCH) model, and the findings showed that there is a general tendency for oil price shocks to be around 1.83 percent bigger in declining oil prices than in rising prices. The asymmetric shocks to the price of oil completely dominate positive leverage. In order to minimize the economy's reliance on crude oil exports as its main export (and source of revenue), measures for export promotion and diversification should be put in place.

Dynamic covariance estimation via predictive Wishart process with an application on brain connectivity estimation

Modeling the complex dependence in multivariate time series data is a fundamental problem in statistics and machine learning. Traditionally, the task has been approached with methods such as multivariate autoregressive models and multivariate generalized autoregressive conditional heteroskedasticity models, and Gaussian process based methods are recently becoming popular by leveraging the flexibility of non-parametric learning. However, few methods exist that directly model the dynamics of the covariance matrices except generalized Wishart process (GWP), and even the generalized Wishart process is limited with applications on small dataset due to the extremely high computational capacity induced by multiple Gaussian processes. In this regard, a novel stochastic process named as Predictive Wishart Process (PWP) is proposed, which provides a collection of positive semi-definite random matrices indexed by input variables. The PWP projects process realizations of GWP to a lower dimensional subspace to efficiently estimate every GWP. The theoretical properties of it are examined, and both Bayesian inference and efficient variational expectation maximization are explored in relation to it. Moreover, the PWP is empirically tested on synthetically generated time-series data to validate competitive reconstructive performance and efficient predictive performance, and applied on a large-scale real functional magnetic resonance imaging (fMRI) dataset from Human Connectome Project (HCP) to demonstrate its practicality. A thorough statistical analysis with visualizations is conducted on the brain connectivity, and also a PWP-based multi-task learning framework is proposed to extract meaningful features from individual fMRIs.

Global climate change and commodity markets: A hedging perspective

AbstractThis paper aims to measure the tail‐risk dependence between climate change and commodity futures markets. We utilize Morgan Stanley Capital International Climate Change Index (CCI) to serve as a proxy indicator of the stock market climate change for examining the tail‐risk spillover effect among 16 major commodity futures. Using GARCH–Copula–CoVaR framework, we show that extreme climate change has a significant tail‐risk spillover effect on commodity futures markets, in which agricultural and energy futures are affected by climate change most. We then adopt the copula‐based generalized autoregressive conditional heteroskedasticity (GARCH) model to compute the optimal hedge ratio among each pair of CCI with commodity futures, which performs well economic advantages for all 16 commodities. Compared with hedging against commodity index, CCI has a more significant hedging effect on metal and energy subindex futures. Therefore, we finally suggest an effective hedging portfolio composed of CCI and subindex futures in a significant economic sense.