Historically the indices have been FW, dating back to January 1990, but the new AW methodology (same constituents, dating back to 2008) warrants a look at the differences between the two, and which set might perhaps be suitable for use by us and our clients.

The figures for the sub-strategy indices tell the same story: This large skew towards the very large funds in the AW indices means that although theoretically the AW indices could be more representative of the ‘Blue-Chip’ universe that market participants have typically monitored, they will be unduly influenced by the performance of a small number of funds.

Other Considerations In addition to the two main considerations above, there are other issues to take in to account.

; WITH p AS ( /* add your query here */ ) SELECT a.agent_name, a.date, a. ID2 WHERE b.agent ='Y' AND (BETWEEN DATEADD(wk, DATEDIFF(wk, 0, Get Date()),0) AND DATEADD(wk, DATEDIFF(wk, 0, GETDATE()),6)) DECLARE @tbl TABLE(agent_name VARCHAR(10) ,[date] date, Id INT) INSERT @tbl SELECT 'ALEX' , '2015-05-25' , 13 UNION SELECT 'ALEX' , '2015-05-22' , 13 UNION SELECT 'ALICE' , '2015-05-24' , 10 UNION SELECT 'ALICE' , '2015-05-26' , 10 SELECT * FROM @tbl -- This period can be changed DECLARE @Start Date date = '01/01/2015' DECLARE @End Date date = '01/01/2016' ; WITH Numbers (Number) AS ( SELECT number FROM master..spt_values WHERE [type] = 'P' ) ,first Day Of Week (f WDate) AS ( SELECT DATEADD(DAY, n.

ID FROM (SELECT *, ROW_NUMBER() OVER (PARTITION BY agent_name, ID, DATEPART(WEEK, [date]) ORDER BY [date] DESC) seq FROM p ) a WHERE seq = 1 SELECT a.agent_name ,,a.

For instance to try and limit survivorship bias, when a fund is removed from the index (either because of liquidation, manager requested removal, or not satisfying criteria for constituency) then the past performance will always remain in the index up until the point of removal.

Similarly for trying to avoid backfill bias, new funds added to an index will not affect the finalised past performance numbers of said index.

Generally these indices are weighted in two different ways – equal weighted (also called fund weighted), and asset weighted, where the constituents are weighted according to their assets under management ().

There are many providers of hedge fund indices in the marketplace today, and the next section will discuss some of them and Aon’s preferences. The list of providers is a lengthy one; some of the more well-known names include Barclay Hedge, Credit Suisse Hedge Index, Eurekahedge, Hedge Fund Intelligence (HFI), HFN Indices, Hedge Fund Research (HFR), Morningstar, and Societe General (formerly Newedge).

This relatively large differential in 2016 was part of the motivation for this study; typically the more well-known managers in the universe are the ones with the larger asset bases, and there was significant dispersion witnessed by industry participants between these managers and the related FW index numbers.

In terms of correlation between the FW and AW indices, over the period from January 2008 – December 2016 the correlation between the Composite Indices was 0.93.

Similarly for each of the sub-strategy indices the correlation remained above 0.90 with the exception of Macro at 0.86.