Skip to content

Getting Started

This guide shows you how to install voiage and run your first Value of Information analysis.

Installation

Using pip (Published Release)

pip install voiage

Development Installation

For the latest development version with all extras:

git clone https://github.com/edithatogo/voiage.git
cd voiage
uv sync --extra dev

Prerequisites

Python 3.12–3.14
pip and setuptools (included with Python)

We recommend using a virtual environment (venv or conda) to manage dependencies.

Quick Start: EVPI

Expected Value of Perfect Information (EVPI) is the maximum amount one should be willing to pay to eliminate all uncertainty.

import numpy as np
from voiage.analysis import DecisionAnalysis
from voiage.schema import ValueArray

# Sample net benefit data: 3 simulations x 2 interventions
values = np.array([
    [20000.0, 25000.0],
    [21000.0, 24800.0],
    [20500.0, 25250.0],
])

value_array = ValueArray.from_numpy(values, ["Standard care", "New treatment"])
analysis = DecisionAnalysis(nb_array=value_array)

print(f"The per-decision EVPI is: {analysis.evpi():.2f}")

# Population-level EVPI
population_evpi = analysis.evpi(
    population=100000,
    time_horizon=10,
    discount_rate=0.03,
)
print(f"The population-level EVPI is: {population_evpi:.2f}")

Quick Start: ENBS

Once you have an EVSI result and know the cost of research, calculate the Expected Net Benefit of Sampling (ENBS):

from voiage.methods.sample_information import enbs

evsi_result = 500000.0
cost_of_research = 150000.0

enbs_value = enbs(evsi_result, cost_of_research)
print(f"The ENBS is: {enbs_value:.2f}")

if enbs_value > 0:
    print("The research is potentially worthwhile.")
else:
    print("The research may not be worthwhile.")

Next Steps

Introduction to voiage — learn about VOI concepts
User Guide — comprehensive usage instructions
API Reference — detailed API documentation
Examples — runnable notebook tutorials

Available Tutorial Notebooks