David R. Montalván Hernández
Contact
📧 davidricardo888 :at: gmail.com
Languages
Spanish: Native speaker.
English: Professional working proficiency.
💼 Work experience
Lecturer
Instituto Tecnológico Autónomo de México (ITAM), Mexico City, Mexico – (Oct 2019 - June 2020)
Lecturer for a diploma program in Data Science and Machine Learning, specifically tailored for finance professionals. Delivered in-depth courses that covered essential topics, including Python programming, linear algebra, probability, and statistics.
Emphasized hands-on experience with data science and machine learning tools such as scikit-learn, pandas, PyTorch, and TensorFlow.
Guided students through real-world applications, empowering them with the skills to leverage these technologies effectively in the financial sector. Fostered a dynamic and interactive learning environment, ensuring a thorough understanding of complex concepts and their practical implications in finance.
Economic Data Analyst
Bloomberg L.P., Mexico City, Mexico – (Aug 2015 - Aug 2017)
- Streamlined and automated data acquisition processes using SQL, PySpark, and various APIs, efficiently gathering macroeconomic statistics for Latin American countries.
- Maintained constant communication with clients, addressing their questions and providing insights regarding economic indicators published by Bloomberg.
- Partnered closely with the news team to develop compelling press notes and insightful analyses based on economic data.
- Designed and implemented robust data validation processes, ensuring high accuracy and consistency of information.
Stock Indexes Analyst
Mexican Stock Exchange, Mexico City, Mexico – (Jul 2013 - Feb 2015)
I managed the maintenance and enhancement of all stock indexes for the Mexican Stock Exchange, including the Índice de Precios y Cotizaciones (Price and Quotations Index, Bloomberg: MEXBOL). My key achievements included:
- Automating Index Constituents Selection: Streamlined and automated the processes for selecting index constituents, improving efficiency and accuracy.
- Dynamic Index Updates: Ensured indexes were up-to-date by incorporating all significant economic events, such as stock splits, buybacks, and dividends, maintaining their relevance and precision.
- Client Engagement: Provided expert support to clients, addressing inquiries about index construction and calculation methodologies, and offering tailored solutions.
- Innovative Index Creation: Designed and launched new stock indexes to address evolving market demands and opportunities.
In addition, I developed and implemented advanced optimization algorithms in MATLAB for the Bursa Optimo Index (S&P/BMV Bursa Optimo Index), enhancing the performance and reliability of the index.
Teacher Assistant
Universidad Nacional Autónoma de México (UNAM), Mexico City, Mexico – (Aug 2012 - Jun 2013)
As a Teaching Assistant for advanced undergraduate courses in risk theory, I supported the instruction and facilitated learning across key topics, including:
- Ruin Process and Probabilities: In-depth coverage of the ruin process and ruin probabilities, with a focus on the compound Poisson process.
- Insurance Premium Principles: Instruction on the foundational principles used to determine and establish insurance premiums.
- Risk Measures: Comprehensive exploration of risk measures, including Value at Risk (VaR), Conditional Value at Risk (CVaR), and Tail Value at Risk (TVaR).
My role involved assisting students with complex concepts, providing practical examples, and ensuring a thorough understanding of risk theory applications.
🎓 Education
PhD Candidate in Computer Science
Uncertainty in Artificial Intelligence, Eindhoven University of Technology, Eindhoven, The Netherlands - (Nov 2020 - current)
My research centers on uncertainty quantification for tractable deep generative probabilistic models, with a particular focus on using imprecise probabilities to enhance the accuracy and reliability of uncertainty measurements. Broadly, I am exploring innovative alternatives to traditional Bayesian and frequentist probability theories to better handle epistemic uncertainty. This includes developing and applying methods based on imprecise probabilities, which provide a more flexible and robust framework for uncertainty quantification.
In addition to theoretical advancements, I am actively seeking financial applications for these novel approaches. By integrating imprecise probabilities into financial models, I aim to improve risk assessment, forecasting, and decision-making processes in the finance sector. My work contributes to bridging the gap between cutting-edge probabilistic theories and practical financial applications, ultimately enhancing the robustness and reliability of financial analyses and predictions.
Master's Degree in Computer Science
Computer Research Center, Instituto Politécnico Nacional (IPN), Mexico City, Mexico - (Aug 2017 - Nov 2019)
In my master’s I was part of the Artificial Intelligence Laboratory.
Thesis: Automated learning of trading rules for the stock market
Bachelor's Degree in Actuarial Science
Faculty of Sciences, Universidad Nacional Autónoma de México (UNAM), Mexico City, Mexico - (Aug 2008 - Jul 2012)
During my bachelor’s studies I focused on financial derivatives theory.
Thesis: Merton's jump diffusion model for pricing European options: A martingale approach.
🛠️ Skills and tools
Python
I have several years of working experience with mainstream packages used in machine learning and data science such as
- Tensorflow
- Pytorch
- Numpy
- Scipy
- Scikit-learn
- Pandas
- PySpark
R
Working experience with the tidyverse collection of packages for data science.
Other tools and programming languages
- C++11/14/17/20
- SQL
- Git
- Linux
- Excel
- Bloomberg terminal
- VBA
- AWS
📜Certifications
Deeplearning.AI
Natural Language Processing (Certificate)
This specialization comprehends the following courses:
- Natural Language Processing with Classification and Vector Spaces.
- Natural Language Processing with Probabilistic Models.
- Natural Language Processing with Sequence Models.
- Natural Language Processing with Attention Models.
Tensorflow in practice specialization (Certificate)
This specialization comprehends the following courses:
- Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep Learning.
- Convolutional Neural Networks in TensorFlow.
- Natural Language Processing in TensorFlow.
- Sequences, Time Series and Prediction
John Hopkins University
- The R Programming Environment (Certificate)
- Advanced R Programming (Certificate)
- Building R Packages (Certificate)
📰Publications
- Dempster-Shafer Credal Probabilistic Circuits (8th International Conference on Belief Functions, 2024)
📝Writing
👨🏫 Teaching