1from scipy.stats import norm
2import numpy as np
3
4def simulate_returns(n_scenarios, correlation):
5# Simulate correlated asset returns using a factor model
6systemic_factor = np.random.randn(n_scenarios)
7idio_factor = np.random.randn(n_scenarios)
8
9returns = np.sqrt(correlation) * systemic_factor + \
10np.sqrt(1 - correlation) * idio_factor
11
12return returns
13
14analyze_tail_risk()

From Complexity to Clarity

I build robust quantitative models and data solutions that transform uncertainty into a strategic advantage.

- result = guess(data)

+ insight = model.predict(data)

- if result > 0.5:

+ if insight.confidence > 0.99:

The Idea Behind the Office

Making a good decision often comes down to one simple idea: being willing to change your mind as you learn more.

This isn't just a statistical concept; it's the foundation of modern development. The Agile Manifesto wisely teaches that our understanding of a project evolves. The path forward becomes clearer after each milestone, and the best plan is one that adapts to new evidence.

That's also the core principle of Bayesian thinking, and it's the inspiration behind the name "The Posterior Office."

In statistics, the “posterior” is your refined, smarter viewpoint after you’ve analyzed the data. My work acts like a post office for your information: I take in the messy, complex data from each stage of your journey, sort it, and deliver the one thing that truly matters—a clear, updated understanding you can act on with confidence.

Whether that means building a solid data pipeline, developing a model that predicts future risks, or designing a dashboard that answers your questions at a glance, my goal is always the same: to provide the clarity needed to make the next decision the best one.

Interactive Demos

A few interactive examples from my quantitative toolkit.

Customer Lifetime Value (CLV)

Simulate the potential future revenue from a customer, accounting for churn and spend variability. A key tool for marketing and business strategy.

View Demo →

Bayesian A/B Testing

Input successes and trials for two variants to visualize how probability distributions update with new data.

View Demo →

Credit Portfolio Simulation

Simulate a loan portfolio's loss distribution using the Vasicek model. Adjust shocks and correlations to see the impact.

View Demo →

Let's Build Something Together

Interested in collaborating, consulting, or just want to talk risk engineering? I'd love to hear from you.

Get in Touch

Or find me on