Quantitative Investment - Tilburg Investment Club

About the Department

Systematic Investing, Built from Scratch

The Quantitative Investment department is TIC's hub for data-driven finance. We bring together students passionate about mathematics, statistics, and programming to develop systematic investment strategies.

Our department operates through two specialized teams - the Risk Management Team and the Quantitative Research Team - each bringing a unique perspective to how we approach financial markets.

Members gain hands-on experience with Python, R, and industry-standard tools while working on real-world projects ranging from portfolio optimization to algorithmic trading strategies.

Our Teams

Two Teams, One Mission

Each team brings specialized expertise to build robust, data-driven investment strategies.

Risk Management Team

Identifying, measuring, and mitigating financial risks through quantitative frameworks and stress testing methodologies.

Portfolio risk analysis & VaR modelling
Stress testing & scenario analysis
Correlation & volatility modelling
Drawdown analysis & risk budgeting
Regulatory risk frameworks (Basel)
Monte Carlo simulations

Join this team →

Quantitative Research Team

Developing and backtesting systematic trading strategies using statistical models, machine learning, and alternative data.

Factor-based investing strategies
Machine learning for alpha generation
Time series analysis & forecasting
Backtesting & strategy evaluation
Alternative data exploration
Portfolio optimization algorithms

Join this team →

Skills & Tools

What You'll Learn

Six areas every quant member works through — the concept, the mathematics, and the code we actually run.

Python & R Programming

Build quantitative models using pandas, NumPy, scikit-learn and industry libraries. Python for day-to-day research, R for deeper statistical work.

Rolling volatility

vol₂₁ = σ(r_t-20..t) · √252

In practice

import pandas as pd, numpy as np

returns = prices.pct_change().dropna()
vol_21d = returns.rolling(21).std() * np.sqrt(252)
# 21 trading days ~ 1 month, annualised

Used for Every price series in the portfolio runs through this.

Statistical Analysis

Regression, hypothesis testing, time-series analysis and stochastic calculus. We test assumptions — normality, stationarity, autocorrelation — before acting on a signal.

Beta, market sensitivity

β_i = Cov(R_i, R_m) / Var(R_m)

In practice

import statsmodels.api as sm

X = sm.add_constant(mkt_returns)
model = sm.OLS(stock_returns, X).fit()
beta = model.params['mkt']  # t-stat in .tvalues

Used for Feeds the beta column on the risk tearsheet.

Machine Learning

Supervised and unsupervised learning applied to financial data. Random forests, gradient boosting and neural networks — with heavy emphasis on walk-forward validation to avoid look-ahead bias.

Logistic classifier

P(y=1 | x) = 1 / (1 + e^{−(β₀ + β^Tx)})

In practice

from sklearn.ensemble import GradientBoostingClassifier

# walk-forward: train on the past only
for train, test in TimeSeriesSplit(5).split(X):
    clf.fit(X[train], y[train])

Used for The LSTM leg of our monthly pipeline runs on this.

Risk Modelling

VaR, CVaR, stress testing and modern risk frameworks. We run parametric VaR, historical simulation and Monte Carlo; CVaR captures the tail risk that VaR ignores.

Value at Risk & expected shortfall

VaR_α = −σ · Φ⁻¹(α) · V · CVaR_α = E[L | L ≥ VaR_α]

In practice

from scipy.stats import norm

z = norm.ppf(0.05)
var_95 = -z * sigma * portfolio_value
cvar = losses[losses >= var_95].mean()

Used for Both reports feed the quarterly risk review.

Portfolio Optimisation

From Markowitz to Hierarchical Risk Parity — the mathematics behind allocation. We backtest each approach against simpler benchmarks like equal weight.

Mean–variance objective

max_w w^Tμ − (λ/2) w^TΣw · SR = (R_p − R_f) / σ_p

In practice

from scipy.optimize import minimize

obj = lambda w: -(w @ mu - lam/2 * w @ cov @ w)
cons = ({'type': 'eq',
         'fun': lambda w: w.sum() - 1},)

Used for Sets the target weights the club votes on.

Time Series & Stochastic Processes

ARIMA, GARCH and the stochastic models underpinning pricing and volatility work. We rarely implement Black–Scholes directly, but understanding GBM is prerequisite to serious quant work.

GARCH(1,1) and geometric Brownian motion

σ_t² = ω + αε_t−1² + βσ_t−1² · dS_t = μS_tdt + σS_tdW_t

In practice

from arch import arch_model

res = arch_model(returns, vol='GARCH',
                 p=1, q=1).fit(disp='off')
forecast = res.forecast(horizon=21)

Used for Drives the volatility forecast in the risk module.

Our Process

From Hypothesis to Strategy

Every project follows a structured research process grounded in the scientific method.

Research & Ideation

Identify market anomalies, review academic literature, and form testable investment hypotheses.

Data & Development

Gather and clean financial data, then build quantitative models and algorithms in Python or R.

Backtesting & Analysis

Rigorously test strategies against historical data, analyze performance metrics, and manage risk.

Presentation & Review

Present findings to the department, receive peer feedback, and iterate on strategy refinement.

Research Highlights

Featured Projects

Recent work from our quantitative research and risk management teams.

Research Team

ML-Based Portfolio Optimization

Using reinforcement learning to dynamically rebalance portfolios based on market regimes.

Risk Management

Tail Risk & Stress Testing Framework

Building a comprehensive framework for extreme scenario analysis in equity portfolios.

Research Team

Multi-Factor Strategy Backtesting

Evaluating value, momentum, and quality factors across European equities.

Performance

Quant Fund — Portfolio Growth

Illustrative growth of our Quant portfolio since inception. Members see live numbers in the portal.

Join Us

Ready to think quantitatively?

Whether you're into data science, mathematics, or simply curious about quant finance - there's a place for you in our team.

Apply now → Back to home

Data-Driven Investment Strategies

Systematic Investing, Built from Scratch

Two Teams, One Mission

Risk Management Team

Quantitative Research Team

What You'll Learn

Python & R Programming

Statistical Analysis

Machine Learning

Risk Modelling

Portfolio Optimisation

Time Series & Stochastic Processes

From Hypothesis to Strategy

Research & Ideation

Data & Development

Backtesting & Analysis

Presentation & Review

Featured Projects

ML-Based Portfolio Optimization

Tail Risk & Stress Testing Framework

Multi-Factor Strategy Backtesting

Quant Fund — Portfolio Growth

Indexed Performance

Ready to think quantitatively?