Systematic & Quant Trading
Overview:
At Cascade, Systematic & Quant Trading turns raw market data into disciplined, rules‑based strategies.
Our desk designs algorithms that capture short‑ and medium‑term opportunities across global equities—momentum bursts, mean‑reversion pockets, earnings‑driven volatility, and more. What sets us apart is the tight feedback loop with our Equity Research and M&A teams: corporate‑action calendars, sector deep‑dives, and deal pipelines flow directly into our signal library, helping the models anticipate regime shifts rather than simply react. The result is a research‑backed trading framework that adapts to evolving market micro‑structure while keeping risk tightly controlled.
Ionut Nodis
Director Quantitative Strategies & Trading
“As Director of Quantitative Strategies & Trading at Cascade Research, I’m excited to see students blend rigorous data science with the nuanced insights of our fundamental analysts. By combining algorithmic precision with human‑driven sector and M&A expertise, we’re building strategies that not only back‑test well but also make intuitive sense. Every model we deploy is an opportunity for members to sharpen coding skills, deepen market intuition, and contribute to a live trading book that showcases the power of interdisciplinary collaboration.”
– Ionut Nodis
This report benchmarks the most widely used option‑pricing methods — from the discrete‑time Cox‑Ross‑Rubinstein binomial tree to the continuous‑time Black‑Scholes‑Merton model, and on to numerical integration, Monte Carlo simulation, and Fast Fourier Transform (Carr‑Madan) techniques. Using a common test‑bed (spot = 100, strike = 100, σ = 20 %, r = 5 %, T = 1 yr), the study measures each method’s absolute error versus the Black‑Scholes price and the runtime needed to achieve that accuracy.
Key findings: numerical integration delivers near‑machine‑precision results in micro‑seconds; binomial trees converge reliably but become progressively slower as you add time‑steps; Monte Carlo offers unrivalled flexibility, but its accuracy improves only slowly as you increase the number of simulated paths; meanwhile, FFT can price an entire sheet of strike prices in a single pass, provided you fine‑tune it to control numerical artefacts. The paper closes with a head‑to‑head chart plotting accuracy against computation time, giving quants a clear decision map for method selection in live systems.
This study revisits the Efficient Market Hypothesis (EMH) in all three of Fama’s forms and asks a focused question: does a calendar anomaly—the day‑of‑the‑week effect—still exist on the Bucharest Stock Exchange (BET)? Drawing on daily closing prices for BET, the S&P 500 and the Dow Jones Industrial Average from July 2013 to August 2023, the paper tests whether abnormal returns on specific weekdays point to short‑term market inefficiency. The author begins by surveying the EMH and the literature on calendar effects, then outlines an empirical design that combines OLS dummy regressions with ARCH/GARCH models to control for time‑varying volatility. The initial results confirm a statistically significant Tuesday effect; however, once global‑market seasonality is accounted for, the anomaly fades—suggesting that any apparent inefficiency is imported from broader market cycles rather than generated within Romania’s equity market itself.
Developed as an applied project for the “Fixed‑Income Securities and Derivatives” module at UCL, this Streamlit web tool delivers professional‑grade bond analytics in a compact interface.
Key functions
Pricing & risk metrics – clean price, accrued interest, Macaulay/modified duration, and convexity calculated instantly from face value, coupon, yield‑to‑maturity, settlement, and maturity.
Scenario analysis – one‑factor Hull–White engine generates stochastic short‑rate paths and reports value‑at‑risk and P&L distributions; full simulation grid exportable as CSV.
Interactive visuals – Plotly charts display cash‑flow timelines, yield‑path overlays, and price histograms for quick interpretation.
Technical stack
Python 3 · Streamlit · NumPy/pandas (vectorised) · Plotly
Containerised via Docker; source code released under the MIT licence.
Use cases
Supports Cascade’s fixed‑income research by standardising bond valuation, accelerating stress‑testing, and providing reproducible outputs for reports and presentations.
Research:
Overview:
Cascade Research now operates three internally managed model portfolios—Cascade 1, Cascade 2, and Cascade 3—each aligned with one of our core research strengths. Together they give members hands‑on experience across the fundamental, event‑driven, and systematic spectrum while providing a transparent record of performance for recruiters, sponsors, and faculty advisors.
Loading..........
The Data is Not Available
Governance and Reporting
All three funds follow the same oversight framework: weekly committee reviews, monthly risk reports, and a public quarterly factsheet outlining attribution and compliance notes. Trades are executed in a paper‑trading environment until committee sign‑off, after which fictive capital is deployed under tight drawdown limits.
Why three funds?
Running distinct mandates lets the society apply its research where it is strongest, fundamental modelling, M&A expertise, and quantitative methods, while showing members how different approaches behave through market cycles.
Long – Short Equity
Information: XXX
Performance since Establishment (May 2025)
Strategy Approach:
Strategy Breakdown: XXX
Portfolio Breakdown (in %):
Company A
Weighting – 20%
Company B
Weighting – 20%
Company B
Weighting – 20%
Company D
Weighting – 20%
Long – Macro Strategy
Information: XXX
Performance since Establishment (May 2025)
Strategy Approach:
Strategy Breakdown: XXX
Portfolio Breakdown (in %):
Company A
Weighting – 20%
Company B
Weighting – 20%
Company B
Weighting – 20%
Company D
Weighting – 20%