Mariol Jonuzaj

Quantitative researcher with 7 years of experience spanning high-frequency financial modelling, econometrics and market microstructure. Specialised in limit order book dynamics, short-horizon alpha signals, and statistical modelling of price formation using NASDAQ Level-3 data. Experienced in developing LOB simulation environments to evaluate execution-aware trading strategies under realistic market conditions.

Quantitative Researcher – Market Microstructure & HFT at Royal Holloway, University of London (2023-06 – 2026-05)

• High-Frequency Alpha Research - NASDAQ TotalView-ITCH (LOBSTER)
• Developed short-horizon ML alpha signals across 35 S&P 500 equities using NASDAQ Level-3 data.
• Volume-based decision triggers (every 0.1%, 0.5% and 1% of expected trades) suitable for HF market making.
• Achieved up to 75% directional accuracy at sub-10ms horizons.
• Evaluated strategies under P&L, latency costs, slippage, and bid-ask spread constraints.
• Evaluated feature importance using permutation, identifying Trade Flow Imbalance and Midprice Momentum as primary drivers.
• LOB Exchange & Simulator
• Built LOB simulation environments to stress-test high-frequency trading strategies.
• Limit and market orders were calibrated using NASDAQ L3 data.
• Implemented L3 data reconstruction and matching engine to ensure microstructure fidelity.
• Used simulator to test ML-based algorithms under different regimes.
• Different regimes can be changed based on order flow and liquidity conditions.
• Backend in Python (OOP + AsyncIO).

Econometrician at Centre for Financial Econometrics, Lancaster University (2019-10 – 2023-04)

• Developed Bayesian state-space models for volatility estimation and latent structure analysis in financial markets.
• Implemented stochastic volatility models, Kalman filtering, and MCMC methods for noisy market environments.
• Designed regime detection frameworks (K-means, GMM clustering) for medium/low frequency markets.
• Applied time series dimensionality reduction techniques to high dimensional financial datasets.
• Published research at Journal of the Royal Statistical Society - Series A.

PhD in Economics – Lancaster University Management School (2019-10 – 2023-05)

MSc in Economics – Lancaster University Management School (2018-10 – 2019-09)

MSc in Statistics – AUEB (2016-10 – 2018-09)

BSc in Economics – AUEB (2012-10 – 2016-09)