Skip to content
View agraolmez's full-sized avatar
1 follower · 0 following

Block or report agraolmez

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
agraolmez/README.md

Ağra Emir Ölmez

I'm currently an MS Financial Economics student at Columbia Business School, graduating in May 2028. I completed my BSc in Economics, Management and Computer Science at Bocconi University in July 2026.

I'm interested in quantitative finance and portfolio management. Most of my work here is on equity selection, portfolio construction, and testing investment strategies with Python.

Personal Website · LinkedIn · CV

Selected work

Machine-learning long/short equity strategy

An end-to-end equity-selection project using point-in-time market, fundamental, analyst, options, ownership, and news data. Model scores feed a constrained long/short portfolio with modeled trading, financing, and borrowing costs.

  • Thesis evaluation, Jan 2024–Aug 2025: 14.67% annualized net excess return, 6.32% volatility, and 2.32 Sharpe
  • Post-thesis V6.8 replay, Jan 2024–Jul 2026: 33.67% net CAGR, 2.03 Sharpe, and 111.63% cumulative net return

These are historical research results, not live returns or untouched prospective evidence. The replay addendum separates the later reconstruction from the thesis evaluation and documents its assumptions.

SPY machine-learning strategy

A LightGBM SPY/cash strategy using 17 macroeconomic and technical signals, with walk-forward backtesting and bootstrap tests. The reported 2023–Jun 2025 comparison returned 24.6% annually before trading costs, with a 2.17 Sharpe and 11.4% volatility.

Paper on SSRN

Sector ETF momentum

Ranks 11 S&P 500 sector ETFs by three-month momentum and holds the top three with monthly rebalancing. The historical study returned 13.5% annually, with an 18.3% maximum drawdown and 0.81 beta. It had lower drawdown and market sensitivity than SPY, but did not outperform SPY on return or Sharpe.

Experience

  • Turk Telekom — Data Analytics Intern, Summer 2025: Python alternative-data pipelines, RoketAI customer-query tooling, and MCP prototypes for data retrieval and tool execution
  • Albaraka — Venture Capital Intern, Summer 2025: Early-stage company screening, DCF and comparable-company valuation, and the Toptan SME digital-wholesale pilot
  • Cropto — Quantitative Research Intern, Summer 2024: Python trading and risk models, automated market-data pipelines, and commodity and digital-asset research

Tools

Python (pandas, NumPy, scikit-learn, XGBoost, LightGBM, PyTorch, statsmodels, CVXPY) · SQL · R · Git · Excel

Outside finance

Soccer for 15 years · Music production in Logic Pro · Tennis · Chess.com Elo 1500 · Poker

Pinned Loading

  1. equity-selection-thesis equity-selection-thesis Public

    Machine-learning long/short equity research with point-in-time data, constrained optimization, and audited V6 evidence

    Python

  2. ml-spy-trading ml-spy-trading Public

    Walk-forward LightGBM study of SPY direction, allocation, and benchmark performance

    Python

  3. momentum-strategy momentum-strategy Public

    Monthly sector-ETF momentum research with benchmark, drawdown, alpha, and beta analysis

    Jupyter Notebook

  4. agraolmez.github.io agraolmez.github.io Public

    Personal portfolio for quantitative research, machine learning, and investing

    HTML