Projects

[01] “Russian Bonds Market Microstructure and Yield-to-Maturity Analysis (MOEX)”
Author: Egor Grishin · Year: 2026

Project description. This project is a quantitative research and data-engineering study of the Russian corporate and government bond market based on the Moscow Exchange (MOEX) ISS API. The goal of the project is to construct a transparent and reproducible framework for analyzing yield-to-maturity (YTM), maturity structure, risk segmentation, and relative valuation of fixed-income instruments traded on MOEX.

The project implements a full data pipeline: automated data collection from MOEX, data cleaning and normalization, construction of bond-level features (YTM, years to maturity, coupon characteristics), and classification by official listing levels (1–3). A special focus is placed on government OFZ bonds, which are used to estimate a risk-free yield curve and compute credit spreads for corporate bonds across maturities.

The analytical layer includes descriptive statistics, ranking of bonds by YTM, spread analysis relative to the OFZ curve, maturity-based aggregation, and multiple visualizations. These include scatter plots of YTM versus time to maturity, risk stratification by listing level, and an interactive treemap highlighting the top bonds by yield. The project is designed both for academic exploration and as a practical research tool for fixed-income analysis.

Materials: Project notebook (Google Colab)

[02] “Machine Learning: Foundations and Basics (SAIF 2025)”
Author: Egor Grishin · Year: 2025

Project description. This repository contains homework assignments completed for the course Machine Learning: Foundations and Basics at the Shanghai Advanced Institute of Finance (SAIF). The course focuses on the mathematical and statistical foundations of machine learning, with particular emphasis on quantitative and finance-related applications.

The project covers core topics such as probability and information theory, linear algebra and matrix decompositions, numerical optimization, statistical inference, regression, and both supervised and unsupervised machine learning. It serves as a structured record of technical coursework and reflects the development of theoretical understanding alongside practical implementation.

The repository is implemented primarily in Jupyter Notebook using Python, with supporting libraries including NumPy, Pandas, and scikit-learn. Individual assignments include mathematical derivations, coding exercises, and visualizations, making the repository both an academic archive and a compact portfolio of foundational machine learning work.

Materials: GitHub repository