Begin with a manifest, not a reconstruction claim
Two prices with the same display time can measure different things. One may be a single venue's latest trade while the other summarises a defined observation window across several sources. Before calculating a difference, write down what each figure is intended to represent.
For the reference figure, retain its exact name and variant, quote currency, publication time, observation-window boundaries, methodology version and source reference. Separately record which input trades you actually possess. A list of missing constituents is a useful result; it should not be replaced by invented transactions merely to produce a complete-looking number.
Read methodology descriptions in their dated context
CF Benchmarks' May 2023 BRR explainer describes a London 15:00–16:00 window divided into twelve five-minute partitions, with volume-weighted medians combined into an arithmetic mean. This is a dated example of a published aggregation method. It is not a claim that this article has checked the complete current methodology for every benchmark variant.
The example explains why a final quote and a window-based reference need not be identical. For an actual comparison, obtain the applicable current methodology and its variant-specific parameters. Keep the document version in the manifest instead of carrying historical settings forward without checking them.
A miniature example shows what aggregation changes
Suppose an invented partition contains three trades: price 100 with quantity 1, price 100 with quantity 1, and price 120 with quantity 8. Total quantity is 10. In this teaching example, the cumulative quantity at price 100 is only 2; the weighted-median price is therefore 120. There is no exact halfway tie to resolve.
The ordinary mean of the three prices is 106.67 rounded to two decimals. Their quantity-weighted mean is 116. Those are different operations on the same fictional input. This miniature is not a full BRR calculation, does not reproduce the provider's complete rules and does not represent an observed market price.
| Operation on the fictional input | Result | Limit |
|---|---|---|
| Ordinary mean of three prices | 106.67, rounded | Ignores quantities |
| Quantity-weighted mean | 116 | Different operation from weighted median |
| Weighted median in this non-tied example | 120 | One invented partition, not a full benchmark |
Preserve the reference zone when reading in Singapore
Using the historical London-window description only as a teaching example, 15:00–16:00 on 19 May 2023 in London corresponds to 22:00–23:00 in Singapore. London then uses UTC+01:00 and Singapore UTC+08:00. A Singapore record at 23:30 would fall outside that example window even though its local calendar date matches.
Do not relabel a local 15:00–16:00 interval as the London interval. Store both boundary instants with known offsets, the original zone and the converted Singapore display. Keep the provider's inclusion or exclusion rule for boundary trades as a separate methodology field; this article does not infer that rule from a clock label.
Currency is another independent field. A SGD display quote is not automatically an input to a USD-denominated reference. Any FX conversion you perform belongs in a separate analytical column with its own rate and time evidence; it does not become the administrator's official input.
Say what your evidence can reproduce
Finish with a coverage statement: which venues, trades and intervals are present, which are missing, and which methodology version you checked. Preserve original trade identifiers and timestamps when available. If you have only a summary price, record that limitation rather than describing the calculation as independently reproduced.
This manifest does not authenticate a settlement entry or recommend a position. It can show why two figures should remain separate until their definitions agree. The optional link below leads to this site's separate crypto signals and Telegram updates offer; that offer neither validates your inputs nor supplies missing benchmark data.
Primary sources and scope
- CF Benchmarks: BRR Suitability Analysis, May 2023
Historical description of the BRR input window and partition aggregation. Current variant-specific rules, actual constituent trades and a complete reconstruction are not verified here. Checked 2026-10-21.