Quantile Autoregression Research Articles

Stock returns, quantile autocorrelation, and volatility forecasting

We examine stock return autocorrelation at various quantiles of the returns' distribution and use it to forecast stock return volatility. Our empirical results show that the strength of the autoregression varies across the quantiles of the returns' distribution in terms of both magnitude and persistence. Specifically, the autoregression order and magnitude of the coefficients is lower in the left tail in comparison with the right tail. Additionally, we show that the quantile autoregressive (QAR) framework proposed in this study improves out-of-sample volatility forecasting performance compared to the generalised autoregressive conditional heteroscedasticity (GARCH)-type models and other quantile-based models. We also observe greater outperformance in QAR estimates during periods of financial turmoil. Moreover, the QAR method also explains the stylized ‘leverage effect’ associated with asset returns in the presence of volatility asymmetry.

Dynamic Spatial Network Quantile Autoregression

We propose a dynamic network quantile regression model to investigate the quantile connectedness using a predetermined network information. We extend the existing network quantile autoregression model of Zhu et al. (2019b) by explicitly allowing the contemporaneous network effects and controlling for the common factors across quantiles. To cope with the endogeneity issue due to simultaneous network spillovers, we adopt the instrumental variable quantile regression (IVQR) estimation and derive the consistency and asymptotic normality of the IVQR estimator using the near epoch dependence property of the network process. Via Monte Carlo simulations, we confirm the satisfactory performance of the IVQR estimator across different quantiles under the different network structures. Finally, we demonstrate the usefulness of our proposed approach with an application to the dataset on the stocks traded in NYSE and NASDAQ in 2016.

Selecting between causal and noncausal models with quantile autoregressions

Abstract We propose a model selection criterion to detect purely causal from purely noncausal models in the framework of quantile autoregressions (QAR). We also present asymptotics for the i.i.d. case with regularly varying distributed innovations in QAR. This new modelling perspective is appealing for investigating the presence of bubbles in economic and financial time series, and is an alternative to approximate maximum likelihood methods. We illustrate our analysis using hyperinflation episodes of Latin American countries.

The asymmetric effect of tourism, financial development, and globalization on ecological footprint in Turkey.

With the growing interest among researchers in analyzing the ecological footprint of any country, this study focuses on new dimensions to analyze the long-run and short-run asymmetric impact of tourism, financial development, and globalization on ecological footprint in Turkey by using Quantile Autoregressive Distributed Lag model for the period from 1986 to 2018. Further, the EKC hypothesis was also tested. The results show that tourism, globalization, and financial development are positively and significantly associated with the EFP. This means that the increase in these variables will further increase the ecological footprint in Turkey. The U-shaped EKC curve was found to be valid in Turkey. The results also depict nonlinear and asymmetric association among most of the variables. Hence, based on the results, further research directions and practical implications can be suggested.

A Boosting-based Quantile Autoregressive Tree Model for the COVID-19 Time Series

Analysis and modelling of the daily observations is of the interest for both academic and practical needs during the worst public health crisis in decades In t

Carpool to work: Determinants at the county-level in the United States

Using US county-level data for 2017, this paper adopts a data-driven approach to study the main factors influencing carpooling for home-to-work trips. The potential explanatory variables quantify demographic, situational and judgmental characteristics of the counties. Since the proportion of workers carpooling for their commute is not (spatially) randomly distributed at the county level, the inclusion of spatial effects improves considerably the model's fit. The Spatial Autoregressive models show that eight variables (four related to demographics, three to situational and one to judgmental) do the best job of explaining the rates of carpooling. A Spatial Quantile Autoregression is further applied as a flexible and interpretable method to address the fact that some of the leading variables have varying effects on counties with different levels of carpooling. For instance, our results suggest that the agglomeration effect, measured by an increase in population density, has a gradual change in trend pattern because it encourages ridesharing among counties with low levels of carpooling, whereas it deters shared trips in high intensity carpooling areas. Alternatively, a decrease in car ownership, a variable strongly associated to counties' income, will lead to the largest increase in employees using carpooling for their home-to-work travels, and this relationship do not vary across quantiles.

The impact of shale gas development on the U.S economy: Evidence from a quantile autoregressive distributed lag model

The advent of shale gas in economically and commercially viable scale has changed the dynamics of the U.S energy profile making it a net exporter of natural gas. However, despite the growing importance of shale gas, its impact on the economy of US has not been substantially investigated. This creates gaps in the empirical literature of energy economics with limited policy guides for the policy makers and other relevant stakeholders. This paper therefore contributes to the extant literature by examining the impacts of shale gas on economic expansion in the U.S in an augmented framework that includes capital stock and labour as additional variables. With quarterly dataset that covers the period, 2002Q1 to 2019Q1, we employed Cho et al. (2015) [1]’s Quantile Autoregressive Distributed Lag (QARDL) modelling technique to probe the long run relationships between the variables. The results confirm a long-run significant impact of shale gas and capital stock on economic growth of U.S while labour force is found to be a positive but not significant factor in economic growth of the U.S. Investment in capital infrastructure that can enhance the technology of natural gas extraction and fracking is recommended for full optimization of the shale gas consumption. Concerted efforts should also be made, through research and development, to improve the efficiency of labour force.

Analyzing technology-emissions association in Top-10 polluted MENA countries: How to ascertain sustainable development by quantile modeling approach

This study investigates the relationship between technological progression and ambient air pollution in top-10 polluted Middle East and North African (MENA) countries by using monthly data for the period of 1990–2017. The Quantile cointegration proposed by Xiao (2009), Quantile-on-Quantile regression (QQ) proposed by Sim and Zhou (2015), and Quantile Autoregressive Granger causality developed by Troster et al. (2018) are applied. In particular, we examine to which extent, quantiles of technological progression affect the quantiles of ambient air pollution, by developing separate indicators for both the mentioned aspects using Principal Component Analysis (PCA). Our empirical findings unfold mutual dependence between technological progression and ambient air pollution. Furthermore, the results of Quantile Autoregressive Granger causality test conclude a bidirectional causal relationship between technological progression and ambient air pollution.

Revisiting the Environmental Kuznets Curve in Malaysia: The role of globalization in sustainable environment

The present study examines the presence of Environmental Kuznets Curve (EKC) in Malaysia by investigating impact of globalization (including overall, social, economic and political) on ecological footprint in the Malaysian economy using quarterly data from 1970 to 2018. This research utilizes the Quantile Autoregressive Distributed Lag (QARDL) method. This approach is useful as it depicts the causality patterns through various quantiles of the globalization and environmental degradation by providing more comprehensive explanations of the over-all dependence among the aforementioned variables almost ignored by the traditional approach (for instance ARDL and OLS). The outcomes of QARDL showed that the error correction coefficient is significant with the anticipated negative sign for all quantiles. This outcome provides evidence on the presence of considerable return to the long-run equilibrium association between the factors of interest and environmental degradation. Overall, the results recommended that overall globalization and economic globalization enhance the level of environmental degradation in the long run, however, political and social globalization help to reduce the level of environmental degradation in the long term. Moreover, the results of QARDL also confirmed the existence of inverted U-Shape curve in the Malaysian economy. In the shorter-run period, this study found a significant negative impact of overall globalization and political globalization on environmental degradation. The results recommended that the Malaysian government and policymakers need to give more attention to social and political globalization to achieve the goals of sustainable environment in the long-run equilibrium.

How oil prices, gold prices, uncertainty and risk impact Islamic and conventional stocks? Empirical evidence from QARDL technique

There are shreds of evidence of Islamic securities to behave differently from conventional ones, especially under the influence of certain factors such as oil, gold, economic policy uncertainty, and geopolitical risk. This paper has empirically evaluated such pieces of evidence through Quantile Autoregressive Distributed Lags Error Correction Model. Analysis has been performed on monthly returns from Dow Jones Islamic Market and Dow Jones Conventional Market Indexes for the sample period from January 1997 to July 2019. Results suggest that the Islamic stocks do behave differently from conventional stocks only for the long term in case of oil price influence under bullish market conditions; whereas, under bearish market conditions, economic policy uncertainty causes Islamic securities to behave differently. Hence, investment in Islamic stocks can be used for diversification of conventional securities’ portfolio under specific conditions. For instance, under oil price changes Islamic and conventional securities can diversify risk in bullish market trends; such diversification can also be achieved in case of the bearish market trend under economic policy uncertainty shock. The results of this study are significant for policymakers and investors as this will provide a clear picture to the investors regarding their investment with respect to Islamic or conventional markets. A further new basis will be provided to both speculators and portfolio managers of Islamic and conventional markets.

Revisiting the role of renewable and non-renewable energy consumption on Turkey’s ecological footprint: Evidence from Quantile ARDL approach

The current study re-investigates the impact of renewable and non-renewable energy consumption on Turkey’s ecological footprint. This study applies Quantile Autoregressive Lagged (QARDL) approach for the period of 1965Q1-2017Q4. We further apply Granger-causality in Quantiles to check the causal relationship among the variables. The results of QARDL show that error correction parameter is statistically significant with the expected negative sign for all quantiles which confirm an existence of significant reversion to the long-term equilibrium connection between the related variables and ecological footprint in Turkey. In particular, the outcomes suggested that renewable energy decrease ecological footprint in long-run on each quantile. However, the results of economic growth and non-renewable energy impact positively to ecological footprint in long-short run period at all quantiles. Finally, we tested the Environmental Kuznets Curve (EKC) hypothesis and the results of QARDL confirmed the EKC in Turkey. Furthermore, the findings of causal investigation from Granger-causality in quantiles evident the presence of a bi-directional causal relationship between renewable energy consumption, energy consumption and economic growth with ecological footprint in the Turkish economy.

Investigating the Dynamics of Rainfall and Temperature in Three Selected Districts in Ghana: A Quantile Autoregression (QAR) Model Perspective

In this paper, the quantile autoregressive (QAR) model was employed to estimate and investigate the dynamics of average rainfall and temperature within different quantile levels/points, according to the monthly year-on-year average rainfall and temperature indices from three selected districts in Ghana from January 2000 and February 2018. It was found that both rainfall and temperature series are stationary across the three districts in Ghana. And we also observe a statistically significant inverse association among rainfall as well as temperature across the three selected districts in Ghana. It was evidenced that the estimated autoregressive coefficients across the three districts presume different values alongside the conditional distributions of rainfall and temperature indices. And the unit root test indicates the general or absolute stationarity of the series.

A quantile autoregression analysis of price volatility in agricultural markets

AbstractThis paper investigates the dynamics of agricultural price volatility based on a quantile autoregression (QAR) model. The QAR model provides a flexible representation of the distribution of price and its dynamics. The approach is applied to U.S. wheat and corn markets over the period of 1980–2017. This period is of significant interest as it covers important changes in agricultural policy and increased reliance on markets. The price analysis is conducted conditional on stocks held in the previous period. We show how increasing previous stocks shift the price distribution to the left and decreases the odds of facing price spikes (by shifting down the upper tail of the price distribution). Our analysis also examines the effects of changing public stocks on prices. For both wheat and corn, this reflects changing agricultural policy, contrasting the 1980s (when public stocks were relatively high) with the post‐2005 period (when public stocks became zero). We document how higher public stock ratio during the previous period did not lower the odds of facing price spikes. Applied to the wheat and corn markets, we also uncover evidence of local dynamic instability in the upper tail of the price distribution, suggesting that price instability becomes more pronounced when previous stocks are low.

The dynamics and volatility of prices in multiple markets: a quantile approach

This paper presents an econometric investigation of price dynamics and volatility in multiple markets. The econometric approach relies on a quantile autoregressive (QAR) model and a copula to provide a flexible representation of price dynamics and volatility in related markets. The analysis allows for an arbitrary distribution of prices across markets, nonlinear dynamics and the presence of price cycles. We propose a two-step estimation method to support a consistent estimation of the multivariate price distribution and its evolution over time. The analysis is illustrated in an econometric application to price dynamics in the US pork vertical sector. The application provides new and useful information on the nature of the pork cycle, the linkages between farm price and retail price and the evolving price volatility in this market.

Inflation persistence investigation of china based on quantile autoregressive model

It has been noted that high inflation has a significant threat to many countries developments and the persistence of inflation increases the cost of monetary policy in controlling inflation. Therefore, this paper proposes a unique method to estimate inflation persistence by using quantile autoregressive (QAR) model and to show how various shocks may affect the rate of inflation within different quantile levels. Based on the QAR model, the monthly inflation persistence in China and its dynamics over time are considered in the paper. The data illustrated here are the monthly year-on-year Chinese consumer price index from January 1987 to May 2019. The results indicate that the Chinese inflation series is globally stationary, but exhibits non-stationary behavior in about 44.85% of the observations. Again, it is evidence that the Chinese inflation rate has asymmetric characteristics at different quantile levels in its conditional distribution. Our research indicates that the QAR model outperforms the ordinary least square (OLS) method in terms of heterogeneity association of the inflation dynamics.

Моделирование взаимосвязей между квантилями доходностей российских и иностранных фондовых рынков для оценки рыночных рисков

The article examines cross dependencies in risks of Russian and foreign stock markets. Bivariate quantile autoregression VAR for VaR is used to achieve this goal. It is shown that Russia is a net receiver of external risk. Tail dependencies between markets tend to increase in turbulent times. Information about them helps to better predict market risks. However, for business use less sophisticated risk models are recommended. The results might be applied for risk-management purposes.

Dynamics of the real exchange rate in European emerging economies: Evidence from quantile regression

The sustainability of purchasing power parity (PPP) theory is examined within the quantile autoregression model for the monthly data of the euro and the US dollar-based real exchange rate (RER) in selected European economies (the Czech Republic, Hungary, Poland, Romania, Serbia and Turkey). Period from January 2000 to December 2014 is covered. The application of quantile autoregression model is motivated by the necessity of identifying asymmetric behavior of the RER due to the shocks of different size and sign. The empirical results support to some extent the PPP theory for the euro- and US dollar-based RER in Romania, Serbia, and Turkey. The euro-based RER in Hungary and Poland is also identified to confirm the PPP theory. The dynamics of the RER in the Czech Republic cannot be associated with the PPP validity. The persistence of the euro-based RER is estimated to be more prominent after the depreciation shocks of smaller size.

Persistence of shocks in CDS returns on Croatian bonds: Quantile autoregression approach

The paper aims to examine persistence of shocks in returns on CDS for 5Y Croatian bonds. Based on sample of daily data from January 6, 2004 up until December 13, 2019 the paper evaluated research hypothesis that assumed persistence of shocks in returns on 5Y Croatian bond. To evaluate the research hypothesis, the paper employed quantile autoregression approach and nonparametric time varying autoregreession approach. The empirical results rejected the research hypothesis assuming persistence of shocks in returns on CDS for 5Y Croatian bonds. Based on the results from this paper, returns on CDS from for 5Y Croatian bonds are in line with efficient market hypothesis for endogenous shocks of small magnitude and at the highest level of endogenous shocks. Furthermore, efficient market hypothesis holds during the calm periods, while during the periods with more dynamics in CDS prices the paper suggests profitable strategy for trader and investors. Eventually, the paper contributes to the ongoing discussion regarding efficient market hypothesis while revealing the case of returns on CDS for 5Y Croatian bonds. Furthermore, the paper suggests trading and investment strategies for investors.

The Dynamics and Volatility of Prices in a Vertical Sector

This paper studies the evolving distribution of prices in a vertical sector in the presence of market cycles. It investigates the dynamic linkages between farm and retail prices with an application to the Chinese pork sector. The econometric approach relies on a quantile autoregressive (QAR) model that provides a flexible representation of the dynamics of price volatility. It investigates the presence and nature of price cycles and their implications for price transmission. We document that the marketing margin is more sensitive to changes in consumer income than in feed cost. In addition, rising feed cost increases the amplitude of the cycle while higher consumer income increases the period of the cycle.

On the dynamic instability of Arctic sea ice

The last few decades have seen a significant decline in Arctic sea ice, generating concerns about both its causes and its longer-term implications. In this paper, we introduce an empirical technique to examine the dynamics of Arctic sea ice extent. Using quantile autoregression, we find that the negative effect of atmospheric CO2 is stronger in the upper tail of the ice distribution. We also document that Arctic sea ice dynamics have become more unstable over the last three decades, especially during the summer. The rising summer instability occurs across quantiles, indicating that it is due to the joint effects of rising atmospheric CO2 and nonlinear feedbacks (and not due to outside shocks). While we do not find evidence of “critical slowing”, we see the increasing instability as a cause for concern. We also use the model to predict the evolution of Arctic sea ice extent under alternative CO2 concentration scenarios.