Skip to content

App Demo

Streamlit Application

The Share Buyback Strategy Simulator is built with Streamlit, providing an interactive web interface for running simulations.

Interface Overview

The left sidebar contains all simulation parameters:

  • Simulation Parameters: Price dynamics (S0, volatility, drift, days)
  • Execution Parameters: Buyback settings (USD, duration constraints)
  • Strategy Parameters: Strategy-specific settings (discount, min completion)
  • Run Simulation: Button to execute Monte Carlo simulation

Main Panel

The main area displays:

  1. Performance Distribution: Histogram of performance across all simulations
  2. Duration Distribution: How long each strategy takes to complete
  3. Comparison Table: Side-by-side statistics for all strategies
  4. Example Path: Single execution trace showing daily behavior

Screenshots

Screenshots coming soon

App screenshots will be added in a future update.

Performance Histogram

Shows the distribution of performance (in basis points) across 10,000+ simulations:

  • X-axis: Performance relative to TWAP benchmark
  • Y-axis: Frequency
  • Each color represents a different strategy

Strategy Comparison Table

Column Description
Strategy Strategy name and number
Mean (bps) Average performance
Std (bps) Standard deviation
Min/Max Range of outcomes
Duration Average completion time

Execution Path Visualization

For a single random simulation:

  • Price path: Stock price over time
  • Benchmark: Rolling TWAP
  • USD executed: Cumulative buyback progress
  • Daily amounts: Bar chart of daily execution

Try It Yourself

The app may be deployed on Streamlit Cloud:

Launch App

git clone https://github.com/Digital-AI-Finance/ShareBuybacks.git
cd ShareBuybacks
pip install -r requirements.txt
streamlit run app.py

Download from Releases and run directly.

Interactive Features

Parameter Sliders

All parameters can be adjusted in real-time:

  • Drag sliders to change values
  • See immediate effect on simulation
  • Compare before/after results

Tabs

Switch between different views:

  • Simulation: Run Monte Carlo simulations
  • Example Path: Detailed single execution
  • Explanation: Strategy documentation

Export

Results can be copied or the page saved for reporting.

(c) Joerg Osterrieder 2025