Gengulphus wrote:Fair enough, as your decision about your spreadsheet - I've done my part by telling you that it's unnecessary to use earlier-than-transaction-date unit values to avoid circular unitisation calculations, and it's entirely up to you to make your own value judgement about whether you want to use the information. But the formula I gave ("Nnew = Nold * V/(V-C), where Nold and Nnew are the old and new numbers of units, V is the current value of the portfolio including the added cash, and C is the added cash") is no more of a "mathematical gyration" than any other formula involved in unitisation calculations, and my concern was also to make certain other readers had been given the information that it's unnecessary. And they too can decide for themselves whether to take that information on board or leave it - and indeed I suspect that quite a few will leave not just that information, but all information about how to do unitisation!
Gengulphus
I've just compared your formula with how I do it by hand for my income units for the calculation I would do at the end of this month (using todays portfolio value as my crystal ball is on the blink).
Your formula gets 166388.59 units and my calculation gets 166378.38 units a discrepancy of 0.006%. I was going to say I would have to think about going back and correcting my previous calculations, but having seen how small the error is I don't think I'll bother (and June is the month that the error will be largest as divis are largest).
However as the formula is easy to set up in my spreadsheet I'll use it from now on, thanks.
Edit: the discrepancy is a bit larger (0.77%) if you look at the number of new units bought rather than the total units)