sdk-rfi

Python SDK for the RAND Forecasting Initiative (RFI) API.

RFI (formerly INFER) is a crowdsourced forecasting platform run by RAND Corporation, powered by the Cultivate Labs platform. It provides access to policy-relevant forecasting questions, crowd predictions, and forecaster comments/rationales.

Installation

pip install sdk-rfi
# or
uv add sdk-rfi

Quick Start

from sdk_rfi import Client

# Set environment variables: RFI_EMAIL, RFI_PASSWORD
client = Client()

# List active questions
questions = client.questions.list()
for q in questions.questions:
    print(f"{q.id}: {q.name}")

# Get a specific question with answers and probabilities
question = client.questions.get(1234)
for answer in question.answers or []:
    print(f"  {answer.name}: {answer.probability_formatted}")

# Get individual forecasts for a question
forecasts = client.prediction_sets.list(question_id=1234)
for ps in forecasts.prediction_sets:
    print(f"  {ps.membership_username}: {ps.rationale}")

# Get comments
comments = client.comments.list(commentable_id=1234, commentable_type="Forecast::Question")

Backtesting

All data-fetching methods support cutoff_date for backtesting:

# Get data as it was available on 2025-01-01
questions = client.questions.list(cutoff_date="2025-01-01")
forecasts = client.prediction_sets.list(question_id=1234, cutoff_date="2025-01-01")

Authentication

The RFI API requires OAuth2 authentication. Set these environment variables:

• RFI_EMAIL - Your RFI account email
• RFI_PASSWORD - Your RFI account password

Or pass them directly:

client = Client(email="user@example.com", password="secret")

Available Resources

| Resource | Methods | Backtestable | |----------|---------|-------------| | questions | list(), get(id) | Yes | | prediction_sets | list() | Yes | | comments | list() | Yes |

Data Leakage Notes

• Answer.probability fields reflect the current crowd forecast, not historical
• Use prediction_sets.list(cutoff_date=...) to get individual forecasts before a date
• predictions_count and predictors_count are always current totals