Join our investing community

Analysis of unit prices - what am I doing wrong?

Discussion in 'Managed Funds & Index Funds' started by johnnyb, 18th Apr, 2007.

  1. johnnyb

    johnnyb Well-Known Member

    Joined:
    16th Aug, 2005
    Posts:
    190
    Location:
    Hobart
    Hi All,

    Inspired by Sim's detailed analysis of his managed funds I thought I would do a bit of investigation myself. As an example I downloaded the history of unit prices for the Platinum International fund. This consisted of unit prices for the fund as well as distributions at the end of each financial year.

    To take account of the distributions I simply added them back onto the unit prices. If I look at my Excel plot of unit prices (see attached) it goes from $1 at inception to just under $4 now. But if I look at the chart on the platinum website it goes from 100 to about 650.

    Shouldn't they both be at the same level (ie just under 4, or 400 on their scale). Am I doing something wrong? I get the feeling I've made beginner error #1 :eek:.

    Thanks.

    John.
     

    Attached Files:

  2. DaveA

    DaveA Well-Known Member

    Joined:
    19th Feb, 2007
    Posts:
    617
    Location:
    Sydney, NSW
    they probably have dividend reinvestments, so come 30 june all the money goes into buying more shares so you get the compounding effect...

    so try doing that
     
  3. johnnyb

    johnnyb Well-Known Member

    Joined:
    16th Aug, 2005
    Posts:
    190
    Location:
    Hobart
    OK, I see.

    So rather than adding the distribution back onto the unit price, I leave the unit price as is, but increase the number of units at each distribution by

    (distribution*existing number of units)/unit price

    Is that right?

    John.
     
  4. Simon Hampel

    Simon Hampel Co-founder Staff Member

    Joined:
    9th Jun, 2005
    Posts:
    4,619
    Location:
    Sydney, Australia
    Pretty much yes ... my calculations are based on reinvesting distributions.

    At the end of the distribution period (quarter/half/year), you work out the dollar value of the distribution, and then use the post-distribution unit price (or the price on the first day of the next period) to calculate how many additional units you would be allocated, and then work out the next period based on that new number of units.

    The trick is finding the post-distribution price !! If a fund manager doesn't publish that data, I typically use the pre-distribution price minus the amount of distribution to calculate an estimate of the post-distribution price. It's not entirely accurate, but it's close.

    When you see performance figures reported by managed funds, they are typically based on an initial investment of $10,000 and then they calculate the percentage increase in value based on the additional units from reinvestment and the current unit price.
     
  5. johnnyb

    johnnyb Well-Known Member

    Joined:
    16th Aug, 2005
    Posts:
    190
    Location:
    Hobart
    Thanks Sim.

    Isn't the post-distribution price just the published price after the distribution. If the fund manager is publishing unit prices then how can they not publish the post-distribution price :confused:. Of have I missed something?

    John.
     
  6. Simon Hampel

    Simon Hampel Co-founder Staff Member

    Joined:
    9th Jun, 2005
    Posts:
    4,619
    Location:
    Sydney, Australia
    Most fund managers publish two unit prices for the end of the period (have a look at the historical unit prices for the CFS funds on their website). There is the unit price as of the end of trading on the last day of the period, and then for the same day, there is a second unit price from after the distribution has been calculated (which is lower).

    If (like I suspect Navra does), they calculate redistributions using the unit price as of the end of trading on the first date of the next period, then there's no need to calculate a separate post-distribution price.