Quantify Investment research that keeps its history

methodology/hrp@3 vs methodology/hrp@2

Not comparable

1 contract difference prevents these figures from being read against each other.

Where the boundary falls

3 versions, 2 comparability walls

not comparable not comparable @1 @2 @3 · latest
BoundaryBlocking difference
@1 → @2 rebalance frequency 5B -> 21B why
@2 → @3 weight bound max 1.0 -> 0.25 why

Blocking differences

Field32Why it matters
weight bound max 0.25 1.0 The earlier version could take positions this one cannot (or vice versa), so the two describe different strategies rather than the same strategy tuned differently.

Differences that do not block

Field32
assumptions ['A 25% ceiling is binding often enough to diversify without... ['A 21-business-day rebalance is frequent enough to track...
assumptions_ref ['assumption/adjusted-close-is-restated@1',... ['assumption/adjusted-close-is-restated@1',...
claims_ref ['claim/hrp-outperforms-mvo-out-of-sample@1'] ['claim/hrp-diversifies-without-constraints@1',...
contract {'universe': ['SPY', 'SH', 'TLT', 'TBT', 'LQD', 'DBC', 'GLD',... {'universe': ['SPY', 'SH', 'TLT', 'TBT', 'LQD', 'DBC', 'GLD',...
limitations ['Estimated on daily Adj Close, which is restated over time by the... ['Estimated on daily Adj Close, which is restated over time by the...
objective Allocate across a fixed multi-asset ETF universe using... Allocate across a fixed multi-asset ETF universe using...
params {'correlation_distance': {'value': 'sqrt_half_one_minus_rho',... {'correlation_distance': {'value': 'sqrt_half_one_minus_rho',...
rules [{'id': 'concentration_cap', 'enforced_by':... [{'id': 'min_history', 'enforced_by': 'params.lookback',...

These changed without changing what the methodology promises its consumers.

How comparability could be restored

Performance comparison

Performance comparison unavailable — versions are directly comparable, performance classes are declared, publication permits this surface

Versions are directly comparable 1 blocking contract difference(s)
Performance classes are declared no performance_class attached
Publication permits this surface no publication decision recorded
Historical and forward series would be separated

A performance visual is itself a claim. It renders only when every condition holds — otherwise the chart would assert something the artifacts do not support.

Why methodology/hrp@2 exists

Erratum 2026-07-30-01 established that v1's published performance was computed without an execution lag or transaction costs. Correcting those revealed that v1's 5-business-day rebalance carries turnover the corrected cost model makes material. Lengthened the rebalance interval to 21 business days and added an explicit turnover cap, both of which reduce cost drag. Also lengthened the correlation lookback from 252 to 504 days to stabilise the clustering, which was flipping cluster membership between adjacent rebalances. Comparability: this changes the rebalance frequency in the output contract, so results published against v1 describe a different strategy and must not be linked to v2's.

← back to hrp