Research
“A discovery is said to be an accident meeting a prepared mind.” ― Albert Szent Gyorgyi
Selected Publications:
"Value-at-Risk under Measurement Error," (with Doukali, M., Song, X.), 2023, Accepted at Oxford Bulletin of Economics and Statistics.
"Portfolio Selection Under Non-Gaussianity And Systemic Risk: A Machine Learning Based Forecasting Approach", (with Lin, W.), 2023, Accepted at International Journal of Forecasting.
"Portfolio Selection Under Systemic Risk", (with Lin, W. and Olmo, J.), 2023, Accepted at Journal of Money, Credit and Banking.
"Testing the Eigenvalue Structure of Spot and Integrated Covariance", 2021, Accepted at Journal of Econometrics, (with Dovonon, P., Williams, J.)
''COVID-19 Control and the Economy: Test, Test, Test'', 2021, Accepted at Oxford Bulletin of Economics and Statistics.
"Measuring Nonlinear Granger Causality in Quantile", forthcoming, Journal of Business & Economic Statistics, (with Song, X.)
"The Information Content of Forward Moments", 2019, Journal of Banking and Finance, (with Andreou, P., Kagkadis, A. Philip, D.)
"Measuring Nonlinear Granger Causality in Mean", 2017, Journal of Business & Economic Statistics, (with Song, X.)
"The Equity Premium, the Variance Premium and the Maturity Structure of Uncertainty", 2014, Review of Finance, (with Feunou, B., Fontaine, J.S., Tédongap, R.)
"Nonparametric Estimation and Inference for Conditional Density based Granger Causality Measures", 2014, Journal of Econometrics,. (with Bouezmarni, T., El Gouch A.)
"Measuring High-Frequency Causality Between Returns, Realized Volatility and Implied Volatility", 2012, Journal of Financial Econometrics,. (with Dufour, J-M., Garcia, R.)
"Short and Long Run Causality Measures: Theory and Inference", 2010, Journal of Econometrics, (with Dufour, J-M.)
"A Nonparametric Copula Based Test for Conditional Independence with Applications to Granger Causality", 2012, Journal of Business & Economic Statistics,. (with Bouezmarni, T., Rombouts, J.V.K.)
"Asymptotic Properties of the Bernstein Density Copula for Dependent Data", 2010, Journal of Multivariate Analysis, (with Bouezmarn, T., Rombouts, J.V.K.)
"Analytical Value-at-Risk and Expected Shortfall under Regime Switching", 2009, Finance Research Letters.
Working Papers:
Research on Covid-19
''Copula-Based Estimation of Health Inequality Measures With an Application to COVID-19'', 2022, (with Bouezmarni, T. and Doukali, M.)
''COVID-19 Control and the Economy: Test, Test, Test'', 2021, Accepted at Oxford Bulletin of Economics and Statistics.
See some reactions in the news about my research on Covid-19:
Newspaper articles on Economy versus Testing, Lockdown
iNews: https://inews.co.uk/news/health/coronavirus-mass-testing-reopen-economy-hard-lockdown-448629
Mirage News: https://www.miragenews.com/testing-cheaper-than-lockdown/
Target Publishing: https://www.targetpublishing.com/pdfs/imagjulaug20.pdf
BizEd: https://bized.aacsb.edu/articles/2020/august/tests-versus-lockdowns
Wellbeing News: https://wellbeingnews.co.uk/covid-19/mass-testing-is-the-safest-way-to-reopen-the-economy-and-society/
India EducationDiary: https://indiaeducationdiary.in/mass-testing-is-the-safest-way-to-reopen-the-economy-and-society-and-will-cost-much-less-than-a-hard-lockdown-research-reveals/
Newspaper articles on Education versus Testing
Education Business: https://educationbusinessuk.net/news/25092020/daily-covid-19-testing-schools-vital-keeping-schools-open
The Educator: http://www.the-educator.org/daily-covid-19-testing-in-schools-is-vital-to-save-childrens-education/
ICT For Education : https://www.ictforeducation.co.uk/news/daily-covid-19-testing-in-schools-is-vital-to-save-children-s-education.html
Viralstories: https://viralstories.co.uk/2020/10/13/covid-19-testing-needed-in-schools-durham-university/
Mirage News: https://www.miragenews.com/covid-19-testing-needed-in-schools/
The Independent Schools Magazine: Daily COVID-19 testing in schools ‘vital to save children’s education
Other research:
"Machine Learning Based Portfolio Selection Under Systemic Risk", 2023 (with Lin, W.)
"Value-at-Risk under Measurement Error", 2020, (with Doukali, M., Song, X.) The paper can be watched here:
Talk at SOAS Centre for Global Finance
Talk at Center for econometrics and business analytics
"Portfolio Selection Under Systemic Risk", 2020 (with Lin, W. and Olmo, J.)
Old Papers:
"Parametric Portfolio Policies with Common Volatility Dynamics", 2015, (with Yunus Emre Ergemen)
Abstract: A parametric portfolio policy function is considered that incorporates common stock volatility dynamics to optimally determine portfolio weights. Reducing dimension of the traditional portfolio selection problem significantly, only a number of policy parameters corresponding to first- and second-order characteristics are estimated based on a standard method-of-moments technique. The method, allowing for the calculation of portfolio weight and return statistics, is illustrated with an empirical application to 30 U.S. industries to study the economic activity before and after the recent financial crisis
"Estimating and Forecasting Financial Risk: A Realized Quantile Approach", 2014, (with Fabian Rinnen and Jesus Gonzalo)
Abstract: This paper introduces a novel methodology for measuring risk based on realized quantiles. The main objective is to better estimate and forecast upward and downward quantiles in a simple way. Two estimation techniques are discussed using order statistic and realized volatility-based measures. Effects on both estimation methods of common misspecications associated with financial intraday return data (e.g. microstructure noise, jumps and fat tails) are discussed and examined through Monte Carlo experiments. Further, standard methods (e.g. Ordinary Least Squares, Autoregressive models...) of time series analysis for relating realized quantile to its past and the past of other covariates in univariate dynamic models are considered, with particular focus on problems that may arise from the sequential approach of estimating first the realized quantiles and secondly the univariate model parameters. Subsequently, an application to the S&P 500 return index is conducted. Additionally to estimating the realized quantiles of S&P 500 return index, they are also related to their own past and the past of returns using standard univariate time series techniques. The results from the latter are compared to the ones obtained from a benchmark model given by the quantile regression model of Koenker (2005). It is found that in the presence of jumps there is an advantage to using order statistic estimation rather than relying on realized volatility based estimates and assumed normality when estimating realized quantiles. On the other hand, both estimation techniques are negatively affected by microstructure noise. However, in the application it is shown that order statistic based estimation is to be preferred and the sequential estimation poses no problem. Finally, it is found that there is a gain from using intraday data to estimate realized quantile time series before relating them dynamically, rather than applying a quantile regression on aggregated data. This gain is in terms of identification of regression parameters and forecast performance as well as with respect to the interpretation of the results.