Methodology
How we structure a global index correlation engagement from scoping call to signed memorandum.
A brief, not a black box
Our methodology exists so a risk committee can see how conclusions were reached. Every study states the index list, the observation end date, the rolling windows, and the stress periods agreed with you.
- Scoping
We confirm benchmarks, currency treatment, and the committee or trustee date that constrains delivery. You tell us which crisis weeks matter to your book.
- Series lock
Index identifiers are written down. We note vendor sources and any splices or backfills so later readers know what they are looking at.
- Window design
Standard studies use several rolling lengths plus at least one calm and one stress window. We avoid mixing daily and weekly returns without saying so.
- Cluster reading
Beyond pairwise numbers, we group indices that travel together and call out breaks when a relationship that looked stable suddenly does not.
- Memorandum and walk-through
Findings are written for the people who will sit in the meeting. The analyst who signed the note leads the discussion.
What we will not do
We will not present an all-period average as if it described a crisis week. We will not imply that low correlation guarantees future independence. We will not dress a short sample as a robust regime map.
Where this leads
If the method fits your next committee cycle, open a consultation or request a briefing with your index list attached.