Bidias TIOTSOP

Engineering Cycle — Applied Mathematics

• France's leading school of engineering
• Key Courses: Deep Learning, Stochastic Modeling, Statistical Learning, Operations Research

Preparatory Classes

• Cameroon's leading school of engineering
• Key Courses: Algebra, Calculus, Probability, Electromagnetism

Bachelor's in Mathematics

• Top 10% (out of 100+ students)
• Key Courses: Group Classification Theory, Probability, Measure Theory, Numerical Analysis

Implemented models to improve ESG performance under nominal rate shocks and default risk on private bonds.

• Designed and calibrated the Hull-White model for inflation rate modeling, achieving 100+% improvement in precision
• Integrated the new inflation model into the existing ESG framework, enhancing predictive robustness under stress scenarios
• Developed a Longstaff-Mithal-Neis (LMN) model for credit risk on private bonds (0.01 factor gain in relative precision)

• Modeled a mutual holding situation through mean field games in discrete time
• Analyzed how diversification affects the total system
• Obtained results consistent with the literature: optimal diversification reduces shareholders' risk (variance)

• Implemented time series classification using Dynamic Time Warping and k-NN on exoplanet WASP-126 b photometric measurements
• Developed LSTM and Transformer-based architectures in PyTorch
• Obtained 92% more precise estimates (RMSE) with Transformers

• Modeled portfolios of options (calls/puts), estimating Value-at-Risk using importance sampling, last particle, and splitting methods
• Utilized importance sampling to enrich rare-event scenarios (losses exceeding 90% quantiles)
• Leveraged auto-regressive Markov chain for conditional law simulation

• Built algorithmic trading models using Random Forest and XGBoost across 9 currency pairs
• Achieved Sharpe ratio of 2.96
• Developed feature selection pipeline analyzing 15+ time series features (ATR, ADX, MACD, RSI...)

• Implemented NSGA II algorithms with fast non-dominated sorting
• Results corroborated with Benjamin Doerr's article on exponential iterations for 2+ valued functions
• Obtained 100% Pareto coverage for m=2, 95% for m=4 on m-LOTZ functions

• Mentored students with a strong interest in mathematics, helping them reach their full potential