Financial Econometrics Research Articles

Historical geopolitical risk and the behaviour of stock returns in advanced economies

In this study, we investigate the impact of global geopolitical risk (GPR) of different forms on the economies of advanced countries (G7 and Switzerland). We construct a predictive model, following the approach of Lewellen (2004. “Predicting returns with financial ratios.” Journal of Financial Economics 74 (2): 209–235) and Westerlund and Narayan (2012. “Does the choice of estimator matter when forecasting returns?” Journal of Banking & Finance 36 (9): 2632–2640; 2015. “Testing for Predictability in Conditionally Heteroskedastic Stock Returns.” Journal of Financial Econometrics 13 (2): 342–375), to analyse over a century of GPR indices and stock returns. For robustness, we control for oil price given its strong connection with stock returns of advanced economies and further extend our analysis to out-of-sample predictability. Our findings reveal that GPR is a significant predictor of stock returns in advanced economies, although their stock markets are vulnerable to GPR and particularly suffer greater adverse effects from threats of GPR (such as threats of war and terrorism) than their actual occurrence. Meanwhile, our forecast evaluation results show that the predictive model that accommodates the GPR indices outperforms the benchmark model that ignores the same both in the in-sample and out-of-sample forecast estimates.

Introduction to the 2018 Hal White Memorial Lecture

Every year the Society for Financial Econometrics celebrates the fundamental contributions to econometric theory and practice made by our late colleague and friend Hal White. Central to these celebrations is a lecture presented by an established scholar during a plenary session of the Society’s Annual Conference. This lecture is subsequently published in the journal. The 2018 Hal White Memorial Lecture was given by Joel Hasbrouck based on the paper “Price Discovery in High Resolution.” The lecture, followed by four commentaries and a re-joinder, form the content of this issue of the journal. Joel Hasbrouck’s lecture focuses on a key area in empirical market microstructure which seeks to understand the granular properties of pricing dynamics from the identification of appropriate price innovations and the separation of permanent from transient price impact effects. The lecture addresses this issue by proposing and estimating at resolutions from sub-milliseconds to one second a Vector Error...

Inaccurate Value at Risk Estimations: Bad Modeling or Inappropriate Data?

Forecasting accurate Value-at-Risk (VaR) estimations is a crucial task in applied financial risk management. Even though there have been significant advances in the field of financial econometrics, many crises have been documented throughout the world in the last decades. An explanation for this discrepancy is that many contemporary models are too complex and cannot be easily understood and implemented in the financial industry (Fama in Financ Anal J 51:75–80, 1995; Ross in AIMR conference proceedings, vol. 1993, no. 6, pp. 11–15, Association for Investment Management and Research, 1993). In order to bridge this theory–practice gap, we present a computational method based on the leverage effect. This method allows us to focus on financial theory and remove complexity. Examining the US stock market (2000–2020), we provide empirical evidence that our newly suggested approach, which uses only the most appropriate observation period, significantly increases the accuracy of the Conventional Delta Normal VaR model and generates VaR estimations which are as accurate as those of advanced econometric models, such as GARCH(1,1).

Strong selection consistency of Bayesian vector autoregressive models based on a pseudo-likelihood approach

Vector autoregressive (VAR) models aim to capture linear temporal interdependencies among multiple time series. They have been widely used in macroeconomics and financial econometrics and more recently have found novel applications in functional genomics and neuroscience. These applications have also accentuated the need to investigate the behavior of the VAR model in a high-dimensional regime, which will provide novel insights into the role of temporal dependence for regularized estimates of the models parameters. However, hardly anything is known regarding posterior model selection consistency for Bayesian VAR models in such regimes. In this work, we develop a pseudo-likelihood based Bayesian approach for consistent variable selection in high-dimensional VAR models by considering hierarchical normal priors on the autoregressive coefficients, as well as on the model space. We establish strong selection consistency of the proposed method, namely that the posterior probability assigned to the true underlying VAR model converges to one under high-dimensional scaling where the dimension p of the VAR system grows nearly exponentially with the sample size n. Further, the result is established under mild regularity conditions on the problem parameters. Finally, as a by-product of these results, we also establish strong selection consistency for the sparse high-dimensional linear regression model with serially correlated regressors and errors.

Did COVID-19 Impact the Connectedness Between Green Bonds and Other Financial Markets? Evidence From Time-Frequency Domain With Portfolio Implications

COVID-19 has morphed from a health crisis to an economic crisis that affected the global economy through several channels. This paper aims to study the impact of COVID-19 on the time-frequency connectedness between Green Bonds and other financial assets. Our sample includes the global stock market, bond market, oil, USD index, and two popular hedging alternatives, namely Gold and Bitcoin, from May 2013 to August 2020. First, we apply the methodologies of Diebold and Yilmaz (International Journal of Forecasting, 2012, 28(1), 57–66) and Baruník and Křehlík (Journal of Financial Econometrics, 2018, 16(2), 271–296). Then, we estimate hedge ratios and hedge effectiveness of green bonds for other financial assets. Green bonds are found to have a great weight in the overall network, particularly strongly connected with the USD index and bond index. While the bi-directional relationship with USD persists during COVID, the connectedness with conventional bonds is also strengthened. Notably, we find a weak relationship between Green bonds and Bitcoin, both in the short and long run. As portfolio implications, Gold and USD have the highest hedge ratio, which is confirmed by the hedging effectiveness. In contrast, oil and stocks exhibit the lowest hedging effectiveness. Our findings imply that financial assets might have a heterogeneous relationship with green bonds. Furthermore, despite its infancy, it seems that the role of green bond during a crisis should not be ignored, as it can be a hedger for some assets, while a contagion amplifier during crisis times.

Evaluation of multi-asset investment strategies with digital assets

The drastic growth of the cryptocurrencies market capitalization boosts investigation of their diversification benefits in portfolio construction. In this paper with a set of classical and modern measurement tools, we assess the out-of-sample performance of eight portfolio allocation strategies relative to the naive 1/N rule applied to traditional and crypto-assets investment universe. Evaluated strategies include a range from classical Markowitz rule to the recently introduced LIBRO approach (Trimborn et al. in Journal of Financial Econometrics 1–27, 2019). Furthermore, we also compare three extensions for strategies with respect to input estimators applied. The results show that in the presence of alternative assets, such as cryptocurrencies, mean–variance strategies underperform the benchmark portfolio. In contrast, CVaR optimization tends to outperform the benchmark as well as geometric optimization, although we find a strong dependence of the former’s success on trading costs. Furthermore, we find evidence that liquidity-bounded strategies tend to perform very well. Thus, our findings underscore the non-normal distribution of returns and the necessity to control for liquidity constraints at alternative asset markets.

Time‐frequency dynamics between fear connectedness of stocks and alternative assets

AbstractWe examine linkages between the time‐frequency dynamics of fear (VIX) connectedness across global stock markets and alternative asset markets. To this end, we utilize Diebold and Yilmaz, International Journal of Forecasting, 2012, 28, 57–66; Diebold and Yilmaz, Journal of Econometrics, 2014, 182, 119–134; and Baruník and Křehlík, Journal of Financial Econometrics, 2018, 16, 271–296 connectedness methods to measure the total, short‐, medium‐ and long‐term connectedness. We subsequently use the rolling‐window wavelet correlation framework of Polanco‐Martínez et al. Physica A: Statistical Mechanics and Its Applications, 2018, 490, 1211–1227 to investigate the dynamic relationship structure among global equity market fear and fear in alternative asset markets, namely oil (OVX), gold (GVZ), currency (EVZ), and bond (TYNVI). We find that OVX drives the underlying VIX connectedness of the global equity markets during stress periods. The results also suggest the prospects of a reverse hedge that is driven by GVZ when the demand for gold intensifies. Furthermore, there are structural shifts in EVZ following Brexit, highlighting a negative relationship with the VIX connectedness. Finally, we find that the VIX connectedness typically moves in the opposite direction to the TYNVI. These results are insightful in realizing vulnerable stock markets' behaviour and explaining investors' sentiments and their shifts over time in asset allocations.

Characterizing financial crises using high-frequency data

Recent advances in high-frequency financial econometrics enable us to characterize which components of the data generating processes change in crisis, and which do not. This paper introduces a new statistic which captures large discontinuities in the composition of a given price series. Monte Carlo simulations suggest that this statistic is useful in characterizing the tail behavior across different sample periods. An application to US Treasury market provides evidence consistent with identifying periods of stress via flight-to-cash behavior which results in increased abrupt price falls at the short end of the term structure and decreased negative price jumps at the long end.

Asymptotic theory for QMLE for the real‐time GARCH(1,1) model

We investigate the asymptotic properties of the Gaussian quasi‐maximum‐likelihood estimator (QMLE) for the Real‐time GARCH(1,1) model of Smetanina (2017, Journal of Financial Econometrics, 15(4), 561–601). The developed theory relies on the functional dependence measure and recently developed theory for derivative processes in Dahlhaus etal. (2019, Bernoulli, 25(2), 1013–1044). We prove stationarity and ergodicity of the underlying processes and consistency for the QMLE estimator under mild conditions. Furthermore, under normality of the error term, we also establish asymptotic normality for QMLE, which then becomes MLE, at the usual rate. Finally, in our simulations we show that consistency and asymptotic normality holds for typical sample sizes.

A GARCH Tutorial with R

ABSTRACT Context: modeling volatility is an advanced technique in financial econometrics, with several applications for academic research. Objective: in this tutorial paper, we will address the topic of volatility modeling in R. We will discuss the underlying logic of GARCH models, their representation and estimation process, along with a descriptive example of a real-world application of volatility modeling. Methods: we use a GARCH model to predict how much time it will take, after the latest crisis, for the Ibovespa index to reach its historical peak once again. The empirical data covers the period between years 2000 and 2020, including the 2009 financial crisis and the current 2020’s episode of the COVID-19 pandemic. Conclusion: we find that, according to our GARCH model, Ibovespa is more likely than not to reach its peak once again in one year and four months from June 2020. All data and R code used to produce this tutorial are freely available on the internet and all results can be easily replicated.

Realized Semi(Co)Variation: Signs that All Volatilities are Not Created Equal

I provide a selective review of recent developments in financial econometrics related to measuring, modeling, forecasting and pricing “good” and “bad” volatilities based on realized variation type measures constructed from high-frequency intraday data. An especially appealing feature of the different measures concerns the ease with which they may be calculated empirically, merely involving cross-products of signed, or thresholded, high-frequency returns. I begin by considering univariate semivariation measures, followed by multivariate semicovariation and semibeta measures, before briefly discussing even richer partial (co)variation measures. I focus my discussion on practical uses of the measures emphasizing what I consider to be the most noteworthy empirical findings to date pertaining to volatility forecasting and asset pricing

The uncertainty spillovers of China's economic policy: Evidence from time and frequency domains

AbstractIn this paper, we investigate the uncertainty spillovers among China's four important economic policies (i.e., fiscal, monetary, trade and exchange rate policy) during the period 2000–2020, based on the spillover framework of Diebold and Yilmaz, The Economic Journal, 2009, 119, 158–171 and Diebold and Yilmaz, International Journal of Forecasting, 2012, 28, 57–66. Considering the heterogeneity of shocks in frequency domain, we also use the frequency methodology of Baruník and Křehlík, Journal of Financial Econometrics, 2018, 16, 271–296. Our results are summarized as follows: (a) the uncertainty spillover between China's different economic policies is high (46.11%) and the low‐frequency spillover (24.90%) is higher than the high‐frequency spillover (19.65%); (b) fiscal policy acts as the mainly uncertainty transmitter on high‐frequency, while monetary policy on low‐frequency; (c) the uncertainty spillover increased in the global financial crisis (2007–2009) and the European debt crisis (2010–2011) and (d) we found that the increase in the former is due to long‐term shocks, while the latter is due to short‐term shocks.

Editorial Announcement

Journal Article Editorial Announcement Get access Journal of Financial Econometrics, Volume 18, Issue 4, Fall 2020, Page 653, https://doi.org/10.1093/jjfinec/nbaa035 Published: 17 December 2020

Research constituents, intellectual structure, and collaboration pattern in the Journal of Forecasting: A bibliometric analysis

AbstractThis study provides a retrospective of the Journal of Forecasting (JoF) between 1982 and 2019. Evidence shows a considerable increase in JoF's productivity and influence since 1982. For example, a JoF article, on average, receives more than 18 citations, and contributions from 65 nations appear in the journal. All articles have a forecasting implication, but JoF encourages a wide range of contexts recently featuring value‐at‐risk and financial econometrics. Factors with a positive association with JoF citations include the number of references, article length, article age, special issue, and number of authors.

Empirical likelihood test for the application of swqmele in fitting an arma‐garch model

Fitting an ARMA‐GARCH model has become a common practice in financial econometrics. Because the asymptotic normality of the quasi maximum likelihood estimation (QMLE) requires finite fourth moment for both errors and the sequence itself, self‐weighted quasi maximum exponential likelihood estimation (SWQMELE) has been proposed to reduce the moment constraints but requires the errors to have zero median instead of zero mean. Because changing zero mean to zero median destroys the ARMA‐GARCH structure and has a serious effect on skewed data, this article proposes an efficient empirical likelihood test for zero mean of errors in the application of SWQMELE to ensure that the model still concerns conditional mean. A simulation study confirms the good finite sample performance before applying the test to the US housing price indexes and financial returns for the study of comovement.

Selecting Computational Models for Asset Management: Financial Econometrics versus Machine Learning—Is There a Conflict?

Successful forecasting requires integrating financial theory, market behavior, exploding sources of data, and computational innovation. Building accurate computational models can be achieved by assembling the most comprehensive toolbox. Both financial econometrics and machine learning approaches help to achieve this objective. Machine learning tools provide the ability to make more accurate predictions by accommodating nonlinearities in data, understanding complex interaction among variables, and allowing the use of large, unstructured datasets. The tools of financial econometrics remain critical in answering questions related to inference among the variables describing economic relationships in finance; when properly applied, their role has not diminished with the introduction of machine learning. TOPICS:Big data/machine learning, simulations, statistical methods Key Findings • Successful forecasting requires integration of financial theory, market behavior, exploding sources of data, and computational innovation. • The well-known documented stylized statistical factors associated with financial market variables and their importance in choosing and applying the proper statistical techniques and machine learning algorithms are discussed. • Guidelines are provided for integrating the characteristics of data, inference features of financial econometrics, and prediction capabilities from machine learning.

Introduction to Topics in Modelling Financial and Macroeconomic Time Series

This special issue of Computational Economics features recent and original studies, most of which were presented at the fourth International Workshop in Financial Markets and Nonlinear Dynamics (FMND), organized in Paris in June 2019 ( www.fmnd.fr ). The papers describe and apply recent developments in financial econometrics to model the properties of financial and macroeconomic data. They present different methodologies and provide interesting and original findings.

Estimation with Errors in Variables via the Characteristic Function

Errors in variables in linear regression continue to be a major empirical issue in financial econometrics. We propose a method using the characteristic function (CF) to obtain estimates for linear models with errors in the variables. By assuming that the explanatory variable follows a flexible double-gamma distribution, we obtain closed-form expressions for the analytic CF of the data generating process. Through simulations, we show that our method performs well relative to existing techniques that address EIVs, including the most sophisticated CF-based techniques, the continuum generalized method of moments and nonparametric methods based on empirical cumulants. We further extend our CF technique to a multivariate setting where it continues to produce accurate estimates. We also provide guidance on implementing CF methods and apply the technique to estimating betas based on the capital asset pricing model.

Financial econometrics, mathematics, statistics, and financial technology: an overall view

Based upon my experience in research, teaching, writing textbooks, and editing handbooks and journals, this review paper discusses how financial econometrics, mathematics, statistics, and financial technology can be used in research and teaching for students majoring in quantitative finance. A major portion of this paper discusses essential content of Lee and Lee (Handbook of financial econometrics, mathematics, statistics, and machine learning, World Scientific, Singapore, 2020). Then Lee (From east to west: memoirs of a finance professor on academia, practice, and policy, World Scientific, Singapore, 2017), Lee et al. (Financial econometrics, mathematics and statistics, Springer, New York, 2019a; Machine learning for predicting default of credit card holders and success of kickstarters. Working paper, 2019b), and Lee and Lee (Handbook of financial econometrics and statistics, Springer, New York, 2015) are used to enhance the content of this paper. In addition, important and relevant papers, which have been published in different journals are also used to support the issues discussed in this paper. I have found the applications of financial econometrics, mathematics, statistics, and technology have improved drastically over the last five decades. Therefore, both practitioners and academicians need to update their skills in this area to compete in both financial market and academic research.

A Holistic Approach to Financial Data Science: Data, Technology, and Analytics

The scientific analysis of financial data, both for practical and theoretical purposes, has continuously been an active and evolving area. Traditionally, financial modeling and analysis have primarily relied on financial econometrics, which applies statistical methods to the problems of finance. Recently, advances in analytical and algorithmic methods, powerful technological solutions, and the expansion of the financial data ecosystem have created new opportunities and provided novel and robust ways to solve existing and new problems in finance. To expand the scope of financial econometrics and incorporate these novelties, the field of financial data science (FDS) has emerged. Crucially, FDS adds more interdisciplinary, methodological variety and technological components to the analysis process, thus generating the need for a holistic view in the management of an FDS project. In this article, the author presents a holistic and structured approach to efficiently and effectively manage an FDS project. The main contribution of the article is to provide a comprehensive, financial domain–oriented approach to FDS management that can be used by researchers and practitioners to exploit the opportunities arising from the evolving finance landscape. TOPICS:Statistical methods, simulations, big data/machine learning Key Findings • Financial data science is presented as an emerging field that complements classical financial econometrics. • The financial data ecosystem is discussed in detail, illustrating the types and structures of financial data and their distinguishing features. • A strategy of nine interrelated parts is proposed to efficiently manage a financial data science project. The proposed approach can be used by industry professionals and academic researchers.