Original Curriculum · Quant & Decision Science

Quantitative Thinking

for Decision Systems

Most data practitioners can run an analysis. Fewer can build a decision system. This course closes that gap — from probability fundamentals to financial signals, risk simulation, and real-world constraints — ending with a decision-driven trading simulation you build yourself.

Modules 5 + Capstone
Format Lecture + Exercises
Audience Analysts, PMs & Engineers
Status ↗ Active
Course Overview

What You'll Build

This is a course about turning data into decisions — not just insights. Each module stacks onto the last, culminating in a full simulation you own.

∿
Quantitative Foundations
Probability, distributions, expected value, and the correlation vs causation trap — the mental models every practitioner needs.
◈
Financial Signals
Time series, volatility, and drawdowns — how real market data behaves and how to extract signals from noise.
⇄
Decision Modeling
The crucial leap: from prediction to decision. ML score alone doesn't decide — policy and context do.
⬡
Risk & Simulation
Monte Carlo methods, portfolio stress scenarios, and thinking probabilistically about outcomes before they happen.
Course Structure

The Five Modules

Each module builds on the last. Every concept is paired with a hands-on exercise so the theory lands in practice, not just slides.

Module 01

Foundations of Quant Thinking

Probability, distributions, and expected value — how to reason about uncertain outcomes before they occur
Correlation vs causation — the single most misunderstood concept in data work, and how to catch it
Exercise Simulate a coin toss portfolio — build intuition for variance, expected returns, and why randomness looks like skill over small samples
Module 02

Financial Data & Signals

Time series analysis — trends, seasonality, autocorrelation, and how financial data differs from static datasets
Volatility and drawdowns — measuring risk in price data, rolling windows, and what these tell you about a system's fragility
Exercise Build a simple price signal from raw market data — apply smoothing, identify regimes, and flag entry/exit conditions
Module 03

Modeling for Decisions

Prediction vs decision systems — a model that outputs a score is not a decision; understanding the gap between inference and action
How context and policy transform a signal into a decision — why the same ML score should trigger different actions in different environments
Key Framework ML_score + Policy + Context → Decision
Module 04

Risk & Simulation

Monte Carlo simulation — generating thousands of possible futures to stress-test a system before committing capital or code
Portfolio stress scenarios — tail risk, correlated drawdowns, and how to design systems that survive the scenarios you didn't plan for
Exercise Run portfolio stress scenarios with Monte Carlo — vary assumptions, observe outcome distributions, and identify which parameters drive the most risk
Module 05

Real-World Constraints

Compliance and regulatory limits — how legal and risk constraints shape what a decision system can and cannot do, regardless of what the model recommends
Capital limits and latency — the physical and financial guardrails that turn a theoretical model into a production system
Designing decision systems that degrade gracefully — what happens when data is late, incomplete, or wrong
Final Project

Capstone

Every concept from the five modules converges here. This isn't a test — it's a build.

Capstone Project

Build a Decision-Driven Trading Simulation

Students design and implement an end-to-end trading simulation that applies every layer of the course: a probability-grounded signal, a policy layer that converts scores into actions, risk constraints that enforce position limits and drawdown stops, and a Monte Carlo layer to stress-test the whole system across thousands of market scenarios. The output is a working simulation — not a slide deck.

∿
Signal Engine
A price signal built from time series data, with regime detection and volatility-adjusted sizing.
⇄
Decision Layer
A policy system that converts the ML score into trade decisions under configurable constraints — capital limits, compliance rules, latency budgets.
⬡
Stress Simulation
Monte Carlo stress scenarios across the full simulation — probability of ruin, tail outcomes, and sensitivity analysis across key parameters.
What You'll Leave With

Skills Covered

Decision Systems Monte Carlo Simulation Financial Signal Processing Risk Modeling Probability Theory Distributions & Expected Value Time Series Analysis Volatility & Drawdowns Correlation vs Causation ML Score → Policy → Action Portfolio Stress Testing Compliance-Aware Modeling Latency & Capital Constraints Quantitative Reasoning

Bring this course to your team.

Available as a workshop series, intensive, or self-paced cohort. Get in touch to discuss.

Get in Touch ← All Courses