Hedge Fund Index Methodology

Indices › Methodology

Index methodology

How the PivotalPath indices are built.

Below are the rules that govern index construction.

Request the methodology (PDF)

1 · Construction principles

The indices are rules-based. Constituents are selected by published, objective criteria rather than by editorial judgement, and the rules are applied identically across every strategy.

Three commitments follow from that, and each is testable against the published data rather than taken on trust.

  1. PivotalPath sources funds through institutional partnerships with allocators and prime brokers. It has a direct relationship with every fund on its platform and does not rely on self-reporting.
  2. Historical values are point-in-time and are not revised to flatter.
  3. The rules themselves are published, so a constituent or an allocator can check that they were applied.

2 · Eligibility

Minimum track record
18 months of reported monthly performance.
Minimum assets
$50 million under management.
Assessment date
1 January of each determination year.
Mid-year changes
Funds falling below either threshold during the year remain constituents for that year’s duration. Membership does not change because a month went badly.
Returns basis
Net of all fees, reported in USD.

3 · Avoiding construction biases

Most hedge fund indices depend on voluntary manager reporting, which introduces three distinct distortions. Each is addressed explicitly.

3.1 Selection bias

PivotalPath sources funds through institutional partnerships with allocators and prime brokers. It has a direct relationship with every fund on its platform and does not rely on self-reporting. Peer groups are defined by strategy classification rather than self-description, so a fund cannot present itself into a flattering comparison.

3.2 Survivorship bias

Index values are point-in-time: each historical month reflects the funds available in that month, whether or not those funds still operate today. Closed funds are not removed from history.

The research universe therefore includes funds that no longer exist. That is deliberate, and it is why the universe count is larger than the count of currently operating funds.

3.3 Backfill bias

A fund joining the universe does not have its prior returns retroactively added to published index history. A track record that looked good enough to start reporting does not get to improve the index’s past.

4 · Index hierarchy

Levels as published. Level 4 sub-groups are maintained selectively and are subject to restrained distribution.
Level What it is Weighting
1 Composite Asset-weighted, with an equal-weighted variant
2 Super-strategy AUM-weighted across Level 3. Managed Futures, Multi-Strategy and Volatility Trading are published equal-weighted.
2.1 Complementary Varies. Sits outside the roll-up.
3 Sub-index Equal-weighted across constituents
4 Sub-group Equal-weighted. Restrained distribution.

Complementary indices are constructed across the hierarchy rather than within it: Equity Market Neutral, Equity Quant and Relative Value. Equity Market Neutral is a computed classification — funds with a self-declared low net exposure and a trailing three-year beta to the S&P 500 within ±0.20 — reassessed annually. Equity Market Neutral and Relative Value are equal-weighted across their constituent sets. Equity Quant aggregates the quantitative sub-groups, equal-weighted within each and combined asset-weighted.

5 · Rebalancing and revision

Constituents are fixed at the end of each calendar year for the following year. Index values are published twice monthly.

The current month and the prior six months remain subject to revision as managers finalise reporting. Months older than that are locked.

6 · Confidentiality

Constituent identities are never disclosed. Only distributional statistics — constituent counts, aggregate assets, year-on-year overlap — are published.

This applies without exception, including to index constituents themselves. A manager can be measured against a true peer group without their data ever being identifiable to allocators, to other managers, or to anyone else.

Returns are monthly, net of all fees, in USD. The complete return history is published free under a Creative Commons Attribution licence at github.com/pivotalpath/publicdata, and the full construction methodology — including the treatment of selection, survivorship and backfill bias — is in the index methodology. Past performance does not predict future returns.

Interested in actionable hedge fund data 
and industry–leading research?

We provide transparency for Allocators.

Get Started