Point-in-Time Backtesting

Construct survivorship-bias-free factor portfolios using accepted_at timestamps and PIT universe snapshots.

Python SDKDuckDBMCP Server

The Look-Ahead Bias Problem

Most financial datasets don't tell you WHEN a data point became available. If you use a company's 2023 annual results (filed March 2024) in a January 2024 backtest, you've cheated — that information wasn't available yet. Valuein's accepted_at field records the exact SEC acceptance timestamp for every fact, enabling rigorous PIT backtesting.

Filtering by accepted_at

Always filter by accepted_at <= rebalance_date when selecting signals for a historical portfolio.

Python

from valuein_sdk import ValueinClient, ValueinError

rebalance_date = "2022-01-01"

try:
    with ValueinClient() as client:

        # PIT-safe: each company's latest annual revenue the SEC had
        # accepted by the rebalance date
        revenue_pit = client.run_query(f"""
          SELECT entity_id, period_end, numeric_value AS revenue, accepted_at
          FROM fact
          WHERE standard_concept = 'TotalRevenue'
            AND fiscal_period = 'FY'
            AND accepted_at <= TIMESTAMP '{rebalance_date}'
          QUALIFY ROW_NUMBER() OVER (
            PARTITION BY entity_id ORDER BY period_end DESC, accepted_at DESC
          ) = 1
        """)
        print(revenue_pit.head())

except ValueinError as e:
    print(f"Error: {e}")

Survivorship-Bias-Free Universe

Never use today's S&P 500 list for historical backtests. Use `client.pit_universe()` in the SDK (or the MCP `get_pit_universe` tool) to get the exact constituents at each rebalance date.

Python

from valuein_sdk import ValueinClient, ValueinError

# MCP: get the exact S&P 500 on Jan 1, 2020
# get_pit_universe(as_of_date="2020-01-01", index="SP500")

# Python SDK: pit_universe() reads index_membership (keyed on cik) with the
# half-open effective_date / removal_date window, delisted members included.
try:
    with ValueinClient() as client:
        universe_2020 = client.pit_universe("2020-01-01", index="SP500")
        print(f"S&P 500 universe on 2020-01-01: {len(universe_2020)} companies")
        print(universe_2020[["cik", "company_name", "ticker_at_date"]].head())

except ValueinError as e:
    print(f"Error: {e}")

Full Monthly Rebalance Loop

Combine PIT universe + PIT signals to build a proper factor backtest.

Python

import pandas as pd
from valuein_sdk import ValueinClient, ValueinError

rebalance_dates = pd.date_range("2018-01-01", "2023-12-01", freq="QS")
portfolio_returns = []

try:
    with ValueinClient() as client:
        for date in rebalance_dates:
            date_str = date.strftime("%Y-%m-%d")

            # Step 1: PIT universe on this date (delisted members included)
            universe = client.pit_universe(date_str)
            if universe.empty:
                continue

            # Step 2: latest annual revenue known on this date — fact.entity_id == cik
            ids = ",".join(f"'{cik}'" for cik in universe["cik"])
            signals = client.run_query(f"""
              SELECT entity_id,
                numeric_value AS revenue,
                accepted_at
              FROM fact
              WHERE standard_concept = 'TotalRevenue'
                AND fiscal_period = 'FY'
                AND accepted_at <= TIMESTAMP '{date_str}'
                AND entity_id IN ({ids})
              QUALIFY ROW_NUMBER() OVER (
                PARTITION BY entity_id ORDER BY period_end DESC, accepted_at DESC
              ) = 1
            """)

            # Step 3: rank by revenue (simplified)
            signals["rank"] = signals["revenue"].rank(pct=True)
            top_quintile = signals[signals["rank"] > 0.8]
            portfolio_returns.append({"date": date, "n": len(top_quintile)})

        print(pd.DataFrame(portfolio_returns).tail())

except ValueinError as e:
    print(f"Error: {e}")

Up next

Building a Financial Agent with MCP

Configure Claude or Cursor to query SEC data via the Valuein MCP Server. Write prompts that generate investment research, screen for opportunities, and analyze risk.