PyDecumulate (MVP)

A Streamlit-based Monte Carlo simulator for retirement decumulation planning.

Motivation

As pension systems across Europe and France specifically shift from Defined Benefit (DB) to Defined Contribution (DC) models, individuals increasingly bear market, inflation, and sequencing risks. PyDecumulate is a proof-of-concept built to evaluate the probability of success for different retirement withdrawal strategies.

Core Features

• Lifecycle Modeling: Connects the accumulation and decumulation phases, capturing path-dependent terminal wealth at the point of retirement.
• Strategy Evaluation: Provides a framework to test and compare dynamic, static, and deterministic de-risking trajectories.
• Simulation Engine: Uses Numba JIT-compiled Monte Carlo loops to execute pathwise scenarios.

Planned Features

• Purchasing Power: Tracks real and nominal wealth to account for inflation over long horizons.