Современные технологии в финансах

Бакалавриат 2026/2027

Статус: Курс по выбору (Совместная программа по экономике НИУ ВШЭ и РЭШ)

Кто читает: Отдел сопровождения учебного процесса в бакалавриате

Где читается: Факультет экономических наук

Когда читается: 3-й курс, 3 модуль

Охват аудитории: для своего кампуса

Язык: русский

Кредиты: 3

Контактные часы: 32

Дополнительные материалы в LMS Задать вопрос

Аннотация

This course will cover applications of modern technologies in finance: robo-advisors, credit scoring with alternative data, decentralized lending platforms and crowdfunding, insuretech, modern payment systems, regtech. Реализуется РЭШ Рабочая программа дисциплины доступна по ссылке https://www.nes.ru/sveden/files/zin/Sovremennye_texnologii_v_finansax.docx.pdf

Цель освоения дисциплины

To bridge financial theory and data‑driven methods – equip students with a rigorous understanding of how modern machine learning and statistical techniques can enhance traditional financial models (asset pricing, risk management, portfolio optimization), while appreciating the unique constraints of financial data: non‑stationarity, low signal‑to‑noise ratio, and limited sample sizes. To develop end‑to‑end modeling competence – through hands‑on assignments and case studies, enable students to independently preprocess financial data, engineer relevant features, select and tune appropriate ML models (from regularized regression to recurrent neural networks), and evaluate their performance using both statistical metrics and financial criteria (e.g., Sharpe ratio, maximum drawdown). To foster critical evaluation and diagnostic skills – train students to identify and mitigate common pitfalls in financial machine learning: look‑ahead bias, overfitting, model decay, and regime‑specific instability. Students will learn to compare model outputs against traditional benchmarks (e.g., GARCH, Fama‑French, Black‑Litterman) and make informed trade‑offs between complexity, interpretability, and out‑of‑sample robustness. To embed practical implementation and business awareness – cultivate the ability to translate quantitative findings into actionable investment or risk‑management strategies, considering transaction costs, liquidity constraints, rebalancing frequency, and regulatory requirements (e.g., explainability, stress testing, fair lending). Students will also gain familiarity with modern software frameworks (e.g., Python with scikit‑learn, PyTorch/TensorFlow) and their application in real‑world financial environments.

Планируемые результаты обучения

Apply a broad spectrum of machine learning models – from linear regularized regression and tree‑based ensembles (Random Forest, Boosting) to neural networks (ANN, RNN/LSTM) – to core financial tasks, including asset pricing, credit default prediction, volatility forecasting, and algorithmic trading.