Mohamed El Ghourabi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mohamed El Ghourabi has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
Mohamed Limam
Department of Management Information Systems, Dhofar University, Salalah, Oman
4 shared publicationsImed Gammoudi
3 shared publicationsAmira Dridi
LARODEC, University of Tunis, 41, Avenue de la Liberté, Cité Bouchoucha, Le Bardo 2000 – Tunis, Tunisia
2 shared publications- Aouatef Riahi
Institute of Applied Biological Sciences of Tunis, University Tunis El Manar, Tunis, Tunisia
2 shared publications - Asma Nani
Quantitative Analysis Research Group (Q.U.A.R.G.), University of Manouba, Manouba, Tunisia
2 shared publications - Fedya Telmoudi
University of Tunis, Tunis Business School, BADEM
2 shared publications
Latest Publications
- 1 January 2020
- Published by Inderscience Publishers in International Journal of Entrepreneurship and Small Business
- Vol. 39 (1/2) , 222
Financial stress testing of Tunisian banking sector in worst case scenarios
Financial stress testing (FST) is a key technique for quantifying financial vulnerabilities; it is an important risk management tool. FST should ask which scenarios lead to big loss with a given level of plausibility. However, traditional FSTs are criticised firstly for the plausibility that rose against stress testing and secondly, for being conducted outside the context of an econometric risk model. Hence, the probability of a sever scenario outcome is unknown and many scenarios yet plausible possibilities are ignored. The aim of this paper is to propose a new FST framework for analysing stress scenarios for financial economic stability. Based on worst case scenario optimisation, our approach is able first to identify the stressful periods with transparent plausibility and second to develop a methodology for conducting FST in the context of any financial-economic risk model. Applied to Tunisian economic system data, our proposed framework identifies more harmful scenarios that are equally plausible leading to stress periods not detected by classical methods.
- 1 January 2020
- Published by Inderscience Publishers in International Journal of Entrepreneurship and Small Business
- Vol. 39 (1/2) , 222
Financial stress testing of Tunisian banking sector in worst case scenarios
Financial stress testing (FST) is a key technique for quantifying financial vulnerabilities; it is an important risk management tool. FST should ask which scenarios lead to big loss with a given level of plausibility. However, traditional FSTs are criticised firstly for the plausibility that rose against stress testing and secondly, for being conducted outside the context of an econometric risk model. Hence, the probability of a sever scenario outcome is unknown and many scenarios yet plausible possibilities are ignored. The aim of this paper is to propose a new FST framework for analysing stress scenarios for financial economic stability. Based on worst case scenario optimisation, our approach is able first to identify the stressful periods with transparent plausibility and second to develop a methodology for conducting FST in the context of any financial-economic risk model. Applied to Tunisian economic system data, our proposed framework identifies more harmful scenarios that are equally plausible leading to stress periods not detected by classical methods.
- 30 May 2019
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 50 (11) , 3338-3363
Generalized quasi maximum likelihood estimation for generalized autoregressive score models: simulations and real applications
In this article, we develop a two-step method for conditional Value at Risk (VaR) estimation in the context of the Generalized Autoregressive Score (GAS) models. The first step consists of estimating the volatility parameter by the generalized Quasi Maximum Likelihood Estimator (gQMLE) and in the second step we estimate the theoretical quantile of the innovations by the empirical quantile of the residuals. When the instrumental density q of the gQMLE is not the Gaussian density used in the standard QMLE, or is not the true distribution of the innovations, both the estimations of the volatility and of the quantile are asymptotically biased. In spite of that the two errors counterbalance each other, and we finally get a consistent estimator of the conditional VaR. We establish the asymptotic properties of the gQMLE for GAS models as well as the conditional VaR two-step estimator. Moreover, we discuss how to apply the gQMLE by giving examples of densities distribution and we get worthwhile results.
- 30 May 2019
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 50 (11) , 3338-3363
Generalized quasi maximum likelihood estimation for generalized autoregressive score models: simulations and real applications
In this article, we develop a two-step method for conditional Value at Risk (VaR) estimation in the context of the Generalized Autoregressive Score (GAS) models. The first step consists of estimating the volatility parameter by the generalized Quasi Maximum Likelihood Estimator (gQMLE) and in the second step we estimate the theoretical quantile of the innovations by the empirical quantile of the residuals. When the instrumental density q of the gQMLE is not the Gaussian density used in the standard QMLE, or is not the true distribution of the innovations, both the estimations of the volatility and of the quantile are asymptotically biased. In spite of that the two errors counterbalance each other, and we finally get a consistent estimator of the conditional VaR. We establish the asymptotic properties of the gQMLE for GAS models as well as the conditional VaR two-step estimator. Moreover, we discuss how to apply the gQMLE by giving examples of densities distribution and we get worthwhile results.
- 27 February 2019
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 46 (12) , 2237-2253
Value-at-risk estimation by LS-SVR and FS-LS-SVR based on GAS model
Conditional risk measuring plays an important role in financial regulation and depends on volatility estimation. A new class of parameter models called Generalized Autoregressive Score (GAS) model has been successfully applied for different error's densities and for different problems of time series prediction in particular for volatility modeling and VaR estimation. To improve the estimating accuracy of the GAS model, this study proposed a semi-parametric method, LS-SVR and FS-LS-SVR applied to the GAS model to estimate the conditional VaR. In particular, we fit the GAS(1,1) model to the return series using three different distributions. Then, LS-SVR and FS-LS-SVR approximate the GAS(1,1) model. An empirical research was performed to illustrate the effectiveness of the proposed method. More precisely, the experimental results from four stock indexes returns suggest that using hybrid models, GAS-LS-SVR and GAS-FS-LS-SVR provides improved performances in the VaR estimation.
- 27 February 2019
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 46 (12) , 2237-2253
Value-at-risk estimation by LS-SVR and FS-LS-SVR based on GAS model
Conditional risk measuring plays an important role in financial regulation and depends on volatility estimation. A new class of parameter models called Generalized Autoregressive Score (GAS) model has been successfully applied for different error's densities and for different problems of time series prediction in particular for volatility modeling and VaR estimation. To improve the estimating accuracy of the GAS model, this study proposed a semi-parametric method, LS-SVR and FS-LS-SVR applied to the GAS model to estimate the conditional VaR. In particular, we fit the GAS(1,1) model to the return series using three different distributions. Then, LS-SVR and FS-LS-SVR approximate the GAS(1,1) model. An empirical research was performed to illustrate the effectiveness of the proposed method. More precisely, the experimental results from four stock indexes returns suggest that using hybrid models, GAS-LS-SVR and GAS-FS-LS-SVR provides improved performances in the VaR estimation.
- 30 March 2016
- Published by Springer Nature in Breast Cancer
- Vol. 24 (2) , 238-244
Family history predictors of BRCA1/BRCA2 mutation status among Tunisian breast/ovarian cancer families
Background With the increasing request for BRCA1/BRCA2 mutation tests, several risk models have been developed to predict the presence of mutation in these genes; in this study, we have developed an efficient BRCA genetic testing strategy. Method As first step, to identify predictor variables associated with BRCA status, we have undertaken a cumulative mutation analysis including data from three Tunisian studies. Then, we have developed a logistic regression model for predicting the likelihood of harboring a BRCA mutation. Using receiver operating characteristic curves (ROC), an effective evaluation was performed. A total of 92 Tunisian families were included. Overall, 27 women were positive for BRCA1/BRCA2 deleterious mutations. Results Tow recurrent mutations (c.211dupA and c.5266dupC) explained 76 % of BRCA1-related families and three recurrent mutations (c.1310_1313del, c.1542_1547delAAGA and c.7887_7888insA) explained 90 % of BRCA2-related families. Early age at diagnosis of breast cancer, ovarian cancer, bilateral breast cancer were associated with BRCA1, whereas male breast cancer and four or more breast cancer cases in the family were associated with BRCA2. The area under the receiver operating characteristic curve of the risk score was 0.802 (95 % confidence interval = 0.0699–0. 905). Conclusion Logistic regression reported particular profiles related to BRCA germline mutation carriers in our population, as well as an efficient prediction model that may be a useful tool for increasing the cost-effectiveness of genetic testing strategy.
- 30 March 2016
- Published by Springer Nature in Breast Cancer
- Vol. 24 (2) , 238-244
Family history predictors of BRCA1/BRCA2 mutation status among Tunisian breast/ovarian cancer families
Background With the increasing request for BRCA1/BRCA2 mutation tests, several risk models have been developed to predict the presence of mutation in these genes; in this study, we have developed an efficient BRCA genetic testing strategy. Method As first step, to identify predictor variables associated with BRCA status, we have undertaken a cumulative mutation analysis including data from three Tunisian studies. Then, we have developed a logistic regression model for predicting the likelihood of harboring a BRCA mutation. Using receiver operating characteristic curves (ROC), an effective evaluation was performed. A total of 92 Tunisian families were included. Overall, 27 women were positive for BRCA1/BRCA2 deleterious mutations. Results Tow recurrent mutations (c.211dupA and c.5266dupC) explained 76 % of BRCA1-related families and three recurrent mutations (c.1310_1313del, c.1542_1547delAAGA and c.7887_7888insA) explained 90 % of BRCA2-related families. Early age at diagnosis of breast cancer, ovarian cancer, bilateral breast cancer were associated with BRCA1, whereas male breast cancer and four or more breast cancer cases in the family were associated with BRCA2. The area under the receiver operating characteristic curve of the risk score was 0.802 (95 % confidence interval = 0.0699–0. 905). Conclusion Logistic regression reported particular profiles related to BRCA germline mutation carriers in our population, as well as an efficient prediction model that may be a useful tool for increasing the cost-effectiveness of genetic testing strategy.
- 29 October 2015
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 43 (8) , 1386-1399
On conditional risk estimation considering model risk
Usually, parametric procedures used for conditional variance modelling are associated with model risk. Model risk may affect the volatility and conditional value at risk estimation process either due to estimation or misspecification risks. Hence, non-parametric artificial intelligence models can be considered as alternative models given that they do not rely on an explicit form of the volatility. In this paper, we consider the least-squares support vector regression (LS-SVR), weighted LS-SVR and Fixed size LS-SVR models in order to handle the problem of conditional risk estimation taking into account issues of model risk. A simulation study and a real application show the performance of proposed volatility and VaR models.
- 29 October 2015
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 43 (8) , 1386-1399
On conditional risk estimation considering model risk
Usually, parametric procedures used for conditional variance modelling are associated with model risk. Model risk may affect the volatility and conditional value at risk estimation process either due to estimation or misspecification risks. Hence, non-parametric artificial intelligence models can be considered as alternative models given that they do not rely on an explicit form of the volatility. In this paper, we consider the least-squares support vector regression (LS-SVR), weighted LS-SVR and Fixed size LS-SVR models in order to handle the problem of conditional risk estimation taking into account issues of model risk. A simulation study and a real application show the performance of proposed volatility and VaR models.
- 17 October 2015
- Published by Springer Nature in Breast Cancer
- Vol. 23 (5) , 807-812
Dissimilarity between sporadic, non-BRCA1/2 families and hereditary breast cancer, linked to BRCA genes, in the Tunisian population
Background Most breast cancers (90 %) are sporadic. Only 5–10 % of all cancer cases can be attributed to genetic defects. BRCA genes are strongly incriminated in the hereditary predisposition to the disease. The purpose of our study was to provide more efficient approach to identify pathogenic BRCA mutation carriers and to determine subgroups within the non-BRCA tumor class. Methods Different clinicopathological features, reproductive factors, as well as psychosocial ones were compared in women carrying mutations in the BRCA1/BRCA2 genes (12 cases) with non-BRCA1/2 family tumors (36 cases) and age-matched sporadic cases, unselected for family history (44 cases). Results A BRCA‐related class was yielded based on age at diagnosis (age ≤ 35 years; p = 0.1), molecular subtypes(the triple-negative subtype was predominant: 43 % of cases; p = 0.025) and age at menarche (p = 0.04). Furthermore, a “probably sporadic” class was distinguished using hormonal contraceptive use (through 30–40 years of age; p = 0.039), the number of full-term pregnancies (age ≥40 years; p = 0.01), age at menopause(age > 50 years; p = 0.04) and psychosocial factors (age ≥ 40 years; p = 0.01). However, analysis of non-BRCA1/2 family tumors indicated that they constitute a heterogeneous class, showing few perceptible differences with sporadic group, but distinct from BRCA1/2 tumors. Conclusions In Tunisian population, breast cancer can be classified with a high level of accuracy as sporadic or related to BRCA germline mutations by combining different clinicopathological features and reproductive factors. This can be clinically useful in genetic counseling and decision making for BRCA genetic test.
- 17 October 2015
- Published by Springer Nature in Breast Cancer
- Vol. 23 (5) , 807-812
Dissimilarity between sporadic, non-BRCA1/2 families and hereditary breast cancer, linked to BRCA genes, in the Tunisian population
Background Most breast cancers (90 %) are sporadic. Only 5–10 % of all cancer cases can be attributed to genetic defects. BRCA genes are strongly incriminated in the hereditary predisposition to the disease. The purpose of our study was to provide more efficient approach to identify pathogenic BRCA mutation carriers and to determine subgroups within the non-BRCA tumor class. Methods Different clinicopathological features, reproductive factors, as well as psychosocial ones were compared in women carrying mutations in the BRCA1/BRCA2 genes (12 cases) with non-BRCA1/2 family tumors (36 cases) and age-matched sporadic cases, unselected for family history (44 cases). Results A BRCA‐related class was yielded based on age at diagnosis (age ≤ 35 years; p = 0.1), molecular subtypes(the triple-negative subtype was predominant: 43 % of cases; p = 0.025) and age at menarche (p = 0.04). Furthermore, a “probably sporadic” class was distinguished using hormonal contraceptive use (through 30–40 years of age; p = 0.039), the number of full-term pregnancies (age ≥40 years; p = 0.01), age at menopause(age > 50 years; p = 0.04) and psychosocial factors (age ≥ 40 years; p = 0.01). However, analysis of non-BRCA1/2 family tumors indicated that they constitute a heterogeneous class, showing few perceptible differences with sporadic group, but distinct from BRCA1/2 tumors. Conclusions In Tunisian population, breast cancer can be classified with a high level of accuracy as sporadic or related to BRCA germline mutations by combining different clinicopathological features and reproductive factors. This can be clinically useful in genetic counseling and decision making for BRCA genetic test.
- 8 May 2015
- Published by Wiley in Journal of Time Series Analysis
- Vol. 37 (1) , 46-76
Consistent Estimation of the Value at Risk When the Error Distribution of the Volatility Model is Misspecified
- 8 May 2015
- Published by Wiley in Journal of Time Series Analysis
- Vol. 37 (1) , 46-76
Consistent Estimation of the Value at Risk When the Error Distribution of the Volatility Model is Misspecified
- 9 December 2014
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 42 (4) , 775-788
A new financial stress index model based on support vector regression and control chart
Financial stress index (FSI) is considered to be an important risk management tool to quantify financial vulnerabilities. This paper proposes a new framework based on a hybrid classifier model that integrates rough set theory (RST), FSI, support vector regression (SVR) and a control chart to identify stressed periods. First, the RST method is applied to select variables. The outputs are used as input data for FSI–SVR computation. Empirical analysis is conducted based on monthly FSI of the Federal Reserve Bank of Saint Louis from January 1992 to June 2011. A comparison study is performed between FSI based on the principal component analysis and FSI–SVR. A control chart based on FSI–SVR and extreme value theory is proposed to identify the extremely stressed periods. Our approach identified different stressed periods including internet bubble, subprime crisis and actual financial stress episodes, along with the calmest periods, agreeing with those given by Federal Reserve System reports.
- 9 December 2014
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 42 (4) , 775-788
A new financial stress index model based on support vector regression and control chart
Financial stress index (FSI) is considered to be an important risk management tool to quantify financial vulnerabilities. This paper proposes a new framework based on a hybrid classifier model that integrates rough set theory (RST), FSI, support vector regression (SVR) and a control chart to identify stressed periods. First, the RST method is applied to select variables. The outputs are used as input data for FSI–SVR computation. Empirical analysis is conducted based on monthly FSI of the Federal Reserve Bank of Saint Louis from January 1992 to June 2011. A comparison study is performed between FSI based on the principal component analysis and FSI–SVR. A control chart based on FSI–SVR and extreme value theory is proposed to identify the extremely stressed periods. Our approach identified different stressed periods including internet bubble, subprime crisis and actual financial stress episodes, along with the calmest periods, agreeing with those given by Federal Reserve System reports.
- 24 September 2010
- Published by Taylor & Francis in Quantitative Finance
- Vol. 12 (3) , 329-339
On monitoring financial stress index with extreme value theory
(2012). On monitoring financial stress index with extreme value theory. Quantitative Finance: Vol. 12, No. 3, pp. 329-339.
- 24 September 2010
- Published by Taylor & Francis in Quantitative Finance
- Vol. 12 (3) , 329-339
On monitoring financial stress index with extreme value theory
(2012). On monitoring financial stress index with extreme value theory. Quantitative Finance: Vol. 12, No. 3, pp. 329-339.