Comparing investment performance

Hi all

At the moment, when analysing existing plans we use FE Analytics to build a graph of performance since inception (or earliest data) based on the current portfolio and discrete calendar performance and we do the same against the recommended portfolio.

Recently, a technical specialist from a provider told me that the FCA wants a comparison of 1,3,5 and 10 years performance rather than calendar discrete - but I cannot find anywhere that backs this up. They suggested we show both - what we are doing plus the 1,3,5 and 10 year comparison but I think that's too much!

What does everyone else do?

Comments

  • We tend to use scatter plots which assess volatility against performance rather than performance in isolation, as this gives a more rounded picture of whether the fund offers value for money, or whether too much (or too little) risk is being taken

    That said, it is fairly easy to knock out a chart with these time periods - I think you can get four periods on one page iirc

  • Unless you have the complete history of all trades on both sides for the whole period then it's meaningless.

    I focus my analysis on asset allocation and risk management. I have a custom table and a spreadsheet that builds an asset allocation chart and I'll do a 1 and 3 year scatter as an indicator of what this means in practice. Caveat being that its based on that portfolio backtested over three years without any management
    Benjamin Fabi 
  • The GIPS industry standard and generally used terms are 36 and 60 months, 3 and 5 years, both providing samples size greater than 30 from which to make some inference about performance.

    However, you may also want to include the last 12 months if its from the clients last review.

    Sadly, economics does not provide us with any information over what the best timer period is so I would fall back on using 12m, 36 and 5yrs.

    Performance is either also Time Weighted, which does not account for cash flows or cash flow weighted. Cash flow weighting does cause some issues as the timing of these will impact the outcome, so preferred is the Time Weighted method, but the dietz metod is used for cash flows.

    In general though you should be reporting risk as well in line with the economic principles of efficient portfolio theory.

Sign In or Register to comment.