Quantile Regression Methodology Research Articles

Capital flows-at-risk: Push, pull and the role of policy

We characterise the probability distributions of various types of gross capital flows conditional on information contained in financial asset prices in a panel of emerging market economies, with a focus on ‘tail’ events. Our framework, based on the quantile regression methodology, allows for a separate role of push- and pull-type factors. We find that both push and pull factors have heterogeneous effects across the distributions of gross capital flows, which are most marked in the left tails. We then explore the role of the pre-existing stance of macroprudential and capital flow management policy in shaping probability distributions of capital flows. Macroprudential and capital flow management measures can both be stabilising, leading to less volatile flows and in particular to lower chances of large portfolio outflows. A tighter macroprudential stance can also reduce the sensitivity of portfolio flows to global financial shocks.

STATE-DEPENDENCY IN THE NEXUS BETWEEN DIGITAL FINANCIAL INCLUSION AND ECONOMIC GROWTH IN SUB-SAHARAN AFRICA

Objective: The objective of this study is to investigate state-dependency in the nexus between digital financial inclusion and economic growth in sub-Saharan Africa, with the aim of examining the impact of adopting digital financial technology to reduce financial exclusion and contribute to economic growth. Theoretical Framework: The paper extends the work of Sarma (2008) and Khera et al. (2021), by creating a new financial inclusion index that captures both traditional and digital financial inclusion. Method: The methodology adopted for this research comprises the principal component analysis (PCA) to identify signals in financial inclusion indicators, thereafter financial inclusion indexes for digital and traditional financial services were constructed. The panel quantile regression methodology was employed to analyse the impact of shocks and determine state dependency in the nexus, respectively. Data collection was carried out through secondary sources. Results and Discussion: The result confirms a positive relationship between digital financial inclusion and economic growth with the greatest impact in countries with lower real GDP, confirming state dependency. Research Implications: The findings suggest that policymakers should focus on promoting digital financial inclusion, particularly in countries with low to intermediate levels of income. Originality/Value: This study contributes to the literature by broadening the definition of digital financial inclusion beyond what is currently found in the literature to include vital financial inclusion dimensions such as access to financial services. The relevance and value of this research are further evidenced by the consideration of state-dependency in the relationship between financial inclusion and growth.

Normative institutional pressures and corporate green development: evidence from a quantile regression approach

ABSTRACT This study explores the often-overlooked role of environmental non-governmental organizations (ENGOs) as catalysts for sustainable practices; it unravels the impact of normative institutional pressures (NIPs) on corporate green investment (CGI). By employing the innovative panel quantile regression (PQR) methodology, this research goes beyond conventional approaches to shed light on the non-monotonic association between NIPs and corporate green investment (CGI) among Chinese listed companies (2009 − 2019). Research findings contribute to the literature on institutional forces and corporate environmental sustainability practices. Meanwhile, policymakers may consider developing frameworks that encourage partnerships, knowledge-sharing, and joint initiatives between the corporate sector and ENGOs.

Exchange rates and trade balance dynamics: a quantile regression analysis

ABSTRACT This paper investigates how trade balances respond to fluctuations in exchange rates among the industrialized G10 and emerging BRICS countries, applying a nonlinear quantile regression methodology. By examining the entire conditional quantile process, we reveal diverse impacts of exchange rate dynamics on trade balances across its distribution. Specifically, our analysis highlights distinct effects: advanced economies experience short-term J-curve effects, while emerging markets display price and quantity effects. Furthermore, our study emphasizes the importance of differentiating between bilateral and effective exchange rates. We observe that effective exchange rates do not offer superior insights compared to bilateral rates, particularly evident in BRICS economies due to the dominance of the US dollar in global trade denomination. Our findings underscore crucial policy implications. Reducing reliance on the US dollar emerges as a pivotal strategy, particularly for developing nations. Conversely, for advanced economies, recognizing and leveraging the indirect yet substantial impact of exchange rates, specifically J-curve effects, could offer strategic advantages within policymaking.

A Systemic Contribution and Vulnerability of Non-financial Firms: A Cross Industry Analysis

This paper quantifies the systemic importance of non-financial firms by assessing their contribution and vulnerability to systemic shocks. We apply two firm-specific measures, namely Delta CoVaR (ΔCoVaR) and Marginal Expected Shortfall (MES), to evaluate the systemic risk of 205 non-financial firms listed on the Pakistan Stock Exchange over the period from 2005-2021. We apply quantile regression methodology to quantify (ΔCoVaR). We confirm that a significant number of firms contribute to both system-wide shocks and are vulnerable to systemic risk of the whole system. We find that firms with a high-risk score in one area are not always high-risk in another. Measures of systemic risk vary significantly across time, between firms and industries. Energy and transport industries are top contributors to systemic risk however, tobacco and pharmaceuticals are among the top industries that are vulnerable to systemic risk of the whole system. We conclude that non-financial firms are systemically important and this risk should be mitigated. This research offers significant insights for policymakers and other relevant stakeholders both domestically and internationally. It aims to help policymakers examine their macro-prudential policy, which now solely takes into account financial firms, to limit the risk that can spread throughout the entire system.

Exploring the Effects of Oil and Gas Prices on Industrial Production in Colombia: A Quantile Regression Analysis

This study examines the impact of oil and gas prices on industrial production in Colombia between 1997 and 2022 using the quantile regression methodology (QR) to evaluate seasonality and nonlinear effects on the distribution of the industrial sector, the findings indicate that when industrial production is at low levels, fluctuations in gas prices have minimal impact. Conversely, when industrial production is at medium levels, gas prices have a significant negative impact, while at high levels (above the sixth decile), gas prices have a negative but not significant impact. On the other hand, oil prices generate positive and significant impacts at medium and high production levels. The study concludes that increasing gas production in the country would lead to a reduction in its price, decrease costs, and increase national production, additionally, gas production can be a viable element in energy transition, as it strengthens sectors of the economy other than oil.

노동조합이 임금프리미엄과 임금불평등에 미친 영향

This paper analyzed the effect of labor unions on wage premium and wage inequality using establishment-worker linked panel data which combines workplace panel survey from 2014 to 2019 with employment insurance data.
 First, an unconditional quantile regression methodology was used to estimate how much union wage premiums differed over the entire sample during the period. The result shows that the union raises the wage premiums of the 40th quartile wage group the most and decreases the effect toward the low wage group and especially the high wage group. It seems that union represent the economic interests of the middle and lower wage groups most.
 Estimating the relationship between the inequality index by business establishment and the unionization rate using the fixed effect, the unionization rate showed a statistically significant negative(-) relationship with all inequality indexes such as the Gini coefficient, the Tile index, the Palma ratio, and the p90/p10 quantile.
 The above estimation results imply that labor unions have had a statistically significant impact on inequality reduction in the labor market of at least the sample (business establishment with 30 or more employees), and it is necessary to consider progress in unionization as a way to improve the unequal labor market situation in the future.

Asymmetric effects of monetary policy on non-bank financial intermediation (NBFI) assets: a panel quantile regression approach

PurposeThe purpose of this paper is to fill the momentous gap by explicitly investigating the asymmetric effects of monetary policy (MP) on non-bank financial intermediation (NBFI) assets.Design/methodology/approachThe authors utilized panel data from 29 countries for the period of 2012–2020 and used the quantile regression estimation. In addition to simultaneous quantile regression (SQR), the authors also employ quantile regression with clustered data (Parente and Silva, 2016) and the generalized quantile regression (GQR) method (Powell, 2020).FindingsThe empirical results show a significant heterogeneous impact of MP. While there is a positive relationship between MP and NBFI assets (“waterbed effect”) at lower quantiles of NBFI assets, at middle and higher quantiles, MP has a negative impact on NBFI assets (“search for yield” effect). The authors further find that negative impact strengthens as the quantile levels of NBFI assets rise from mid to high. Findings also reveal that “procyclicality” (except higher quantile) and “institutional demand” hypotheses hold. However, regarding “regulatory arbitrage,” mixed results are observed indicating the impact of Basel III requirements.Originality/valuePrevious empirical studies have concentrated on either the Dynamic Stochastic General Equilibrium (DSGE) framework or conditional mean regression approaches and delivered mixed findings of the MP effects on NBFI. The current paper takes a step toward dealing with this issue by deploying quantile regression methodology, which shows the impact of MP on NBFI at different conditional distributions (quantiles) of NBFI assets instead of just NBFI's conditional mean distribution.

Does resource-richness cause resources curse in financial market? A sustainable development overview for RCEP economies

This study estimates the dynamic association between financial inclusion (FI) and economic growth (GDP) to examine the effect of FI on total natural resource rents (TNRR), GDP, and renewable energy consumption (REC). The current analysis explores the dynamic association between FI, TNRR, Foreign Direct Investment (FDI), and REC using panel data from China, Japan, Australia, Indonesia, and South Korea (RCEP) countries from 2004 to 2019. The Method of Moments Quantile Regression (MMQR) methodology was used to achieve the results. It is believed that by matching the findings of the dynamic panel data, we would be able to assess the overall correlations of variables and derive more significant information. At the 25th to 90th quantiles, GDP increases FI. According to the findings. As for REC and FDI have a favorable impact and can help enhance FI on the 25th to 90th quantiles, respectively. TNRR, in contrast, decreases FI. Furthermore, when implementing RCEP economies' energy, growth policies, and environment centered on empirical data, legislators should analyze the distorted behavior of REC, FDI, and GDP.

Water and traditional asset classes

Data from the most representative water exchange-traded index funds (ETFs) at the worldwide level has been selected to analyze their role in the performance of portfolios that follow the main US and European stock, bond and volatility benchmarks. The empirical analysis has been performed using cross- correlation analysis and Quantile Regression methodology. Our results indicate that water assets can play the roles of diversifier, hedge and safe haven for equity, fixed income and volatility indices, respectively.

Effect of Capital Controls: A Quantile Regression Approach

A neoclassical model reveals that capital controls depress capital flows by adjusting the marginal benefits and costs of additional asset purchases. According to this view, their effectiveness might be dependent on the volume of flows. To study this possibility, we examined the associations between capital flows and controls using the Powell’s (2022) quantile regression methodology. The data used cover 43 emerging market economies between 1995 and 2019. Our results suggest that capital controls are differently associated with capital flows depending on the conditional distributions of capital flows and this implies capital controls could be effective but only when implemented appropriately depending on the volume of inward and outward capital flows. Therefore, the government must carefully and precisely set restrictions depending on the volume of capital flows, which might be an extremely challenging task.

Deep neural networks for the quantile estimation of regional renewable energy production

Wind and solar energy forecasting have become crucial for the inclusion of renewable energy in electrical power systems. Although most works have focused on point prediction, it is currently becoming important to also estimate the forecast uncertainty. With regard to forecasting methods, deep neural networks have shown good performance in many fields. However, the use of these networks for comparative studies of probabilistic forecasts of renewable energies, especially for regional forecasts, has not yet received much attention. The aim of this article is to study the performance of deep networks for estimating multiple conditional quantiles on regional renewable electricity production and compare them with widely used quantile regression methods such as the linear, support vector quantile regression, gradient boosting quantile regression, natural gradient boosting and quantile regression forest methods. A grid of numerical weather prediction variables covers the region of interest. These variables act as the predictors of the regional model. In addition to quantiles, prediction intervals are also constructed, and the models are evaluated using different metrics. These prediction intervals are further improved through an adapted conformalized quantile regression methodology. Overall, the results show that deep networks are the best performing method for both solar and wind energy regions, producing narrow prediction intervals with good coverage.

Capital Structure Decisions and Corporate Performance: Does Firm’s Profitability Matter?

This study aims to investigate the relationship between financing decisions and firm performance. It particularly investigates the heterogeneous effects of capital structure ratios on financial profitability. The study employs the Quantile regression methodology on a sample of 120 non-financial companies listed on Tadawul stock exchange during the period 2017–2020. Financial performance was measured using return on assets, return on equities and Tobin’s Q variables in order to assess accounting and market performance. Data on the various variables is obtained from the companies’ annual reports. Quantile regression results show that debts to equity ratio hamper firms’ performance where as equity financing ratio increases business profitability. Additionally, findings demonstrate that this relationship is nonlinear. Particularly, a debt to equity ratio has a greater negative effect on performance of high-profitable firms. However, the positive effect of the equity financing ratio seems to be higher in high profitable companies than low ones. These results would help managers of non financial firms regarding optimal capital structure decisions. Indeed, managers of non-financial firms could use results of this study as a benchmark to make efficient decisions related to the structure of the capital such as reducing the proportion of debts in the capital and increasing the weight of equity financing. Particularly, enterprises in the early stages of development, with lower profits reflected in return on assets, return on equities, and Tobin’s Q should carefully avoid debt, whereas firms with big earnings are encouraged to raise their capital by issuing new shares in the financial market.

Study of the leading European construction companies using risk factor models

AbstractThis paper aims to study the potential effects of changes in international risk factors on leading European construction companies' returns. The study is conducted on a sample period between January 2000 and December 2019 and applies an extension of the Fama and French five‐factor model (2015) using the quantile regression methodology. Specifically, this research extends the Fama and French (2015) factor model by adding risk factors such as nominal interest rates, momentum and momentum reversal factors (Carhart, 1997) and the traded liquidity factor (Pastor and Stambaugh, 2003). For robustness, this study splits the entire sample period into three sub‐periods: pre‐crisis, crisis, and post‐crisis. As expected, the highest R2 coefficients are observed in the extreme quantiles, that is, in the bullish and bearish market states, revealing a U‐shaped relationship in the explanatory power of this factor model.

Examination of the convergence route to IFRS reporting and disclosure

Value relevance (VR) of earnings and book value of equity is studied in a setting where the International Financial Reporting Standards (IFRS) have been adopted through a convergence and customization route. Quantile regression methodology is applied to level and return models. We find no significant increase in VR of earnings or book equity. Smaller firms show some sensitivity to the change in the regime as compared to the largest set of firms, though accounting metrics overall, help explain the value of larger firms better. We conclude that the convergence route leads to continuous, incremental benefits over the pre-adoption period which pre-empts any significant increase in VR upon IFRS adoption. Gradual convergence with IFRS supported by positive, investor-friendly changes (Roca, 2021) to existing institutional and regulatory frameworks over time, results in better adoption and early, continuous capture of value, though the process itself is long drawn out. More research is needed to test the relevance of alternate metrics in the current technology and intangibles-driven economies (Barth, Li, & McClure, 2021). India’s unique approach to IFRS adoption may hold lessons for all IFRS adopters across the world while responding to new/revised standards in the future. This is the first comprehensive study on the value relevance and information content of the Indian Accounting Standards (Ind AS)

A worldwide tourism-extended Environmental Kuznets Curve. New approaches in a comparative analysis

This work aims to empirically study the exogenous impact of the tourism sector on the environment and the validity of the conventional and tourism-extended Environmental Kuznets Curve (EKC) in 69 countries from America, Europe, and Asia-Oceania between 1995 and 2016. We introduce several interesting approaches in our analysis: the consideration of three heterogeneous regions to isolate the relationship between tourism activity and environmental quality from local conditions, an environmental-legal index to consider the effect of environmental regulations on pollution, and an instrumental quantile regression methodology to avoid possible endogeneity problems. Our results show that tourism activity causes environmental damage, measured by per capita carbon dioxide emissions. We also find evidence suggesting the fulfillment of the conventional and tourism-extended EKC hypotheses for the full sample and American and Asia-Oceania countries, but not European ones. Lastly, an important policy implication emerging from our analysis refers to the importance of regulations in mitigating environmental degradation, especially in high-polluted countries, suggesting the need for governments to take an active role in this matter to ensure sustainable and responsible tourism.

Modeling T20I cricket bowling effectiveness: A quantile regression approach with a Bayesian extension

Bowling effectiveness is a key factor in winning cricket matches. The team captain should decide when to use the right bowler at the right moment so that the team can optimize the outcome of the game. In this study, we investigate the effectiveness of different types of bowlers at different stages of the game, based on the conceded percentage of runs from the innings total, for each over. Bowlers are generally categorized into three types: fast bowlers, medium-fast bowlers, and spinners. In this article, the authors divided the twenty over spell of a T20I match into four stages; namely, Stage 1: overs 1-6 (PowerPlay), Stage 2: overs 7-10, Stage 3: overs 11-15, and Stage 4: overs 16-20. To understand the broad spectrum of the behavior of game variables, a Quantile Regression methodology is used for statistical analysis. Following that, a Bayesian approach to Quantile Regression is undertaken, and it confirms the initial results.

The structure and degree of dependence in government bond markets

Our study provides new evidence on asymmetric dependencies in international government bond markets, by examining bonds from developed, emerging, and frontier countries, using a quantile regression methodology. We find that the dependence structure for emerging and frontier markets significantly changes during financial crisis periods, which we show has important implications for international diversification of investment strategies. Moreover, we also examine in detail stock–bond correlations and uncover several instances of decoupling. In contrast, developed markets exhibit a more stable dependence pattern. In addition, we document that the degree and structure of dependence vary when foreign currencies are hedged or unhedged, and across bond maturity segments.

Quantile regression for modelling the impact of climate in cork growth quantiles in Portugal

The annual growth and the thickness of cork are known to be highly variable between trees located in the same geographical location. Researching how climate variables affect different trees within the same site is a step forward for the management of cork production since current knowledge focusses only on the average tree response. Quantile regression methodology was applied for the first time to a large data set containing measurements of cork growth, sampled in 35 stands across the cork oak distribution area in Portugal. This methodology proved to be useful for testing the hypothesis raised: does climate affect differently the annual cork growth, and ultimately cork thickness of individual trees located in the same stand? Estimating the amount of cork produced by one stand that has the required thickness for the production of natural cork stoppers is essential to support cork oak management. However, no model, before this work, had been developed to provide managers with this information. A downward parabolic relationship between annual cork growth and annual precipitation was determined for all quantiles, with optimum annual average precipitation value ranging from 1103 to 1007 mm. April to August monthly temperatures, spring average temperature or summer average temperature, showed a negative relationship with annual cork growth, in particular for lower quantiles. Maximum annual temperature was shown to negatively affect annual cork thickness, in particular for the trees under the 6th quantile. The ratio between annual precipitation and average temperature, that define the Lang index (LI), showed a downward parabolic relationship with annual cork growth. Best cork growth conditions are found for Lang index values around 60, corresponding for the transition between semi-arid climate and humid climate. The application of the final model developed for estimating cork thickness of an eight years’ cork growth period allowed the prediction and mapping of the percentage of cork suitable for natural cork stopper production. It showed that higher values are expected in the Southern and Central coastal regions and along the Tagus River basin. The Northern coastal and mountain regions, characterised by Lang index values higher to 60 (humid climates), present lower estimated values for the percentage of cork suitable for natural cork stopper production. The estimated values are expected to be reduced under climate change scenarios in the Southern and Central coastal regions.

Quantile regression: prediction of growth and yield for a eucalyptus plantation in northeast Brazil

Regression analysis is a traditional technique to fit equations and predict tree and forest attributes. However, problems may occur when the data show high dispersion around the mean of the regressed variable, limiting the use of traditional methods such as the Ordinary Least Squares (OLS) estimator. Hence, the objectives were to propose a Quantile Regression (QR) methodology to predict tree growth and yield of a forest plantation without using the site index and compare it with the predictive accuracy of the Clutter model. The data came from clonal plantations of Eucalyptus grandis x E. urophylla located in the north of Bahia state (Brazil), and age ranged from 20 to 89-months-old. The set of permanent plots was divided into 50% for model training and the remaining for validation. The volume prediction methods, Clutter and QR, were evaluated using scatter plots of the relative error and graphs of the observed versus predicted volumes, as well as by the following statistics: mean prediction bias in percentage (BIAS %) and root mean square error in percentage (RMSE %). The volume predictions by QR resulted in higher accuracy compared to those obtained by the Clutter model. The QR demonstrated efficiency to predict volumes for different ages of eucalyptus plantations and can also be adapted to other forest areas and species worldwide.