If you just want to calculate an Annualized return (without all the bumpf which follows), go here Suppose we invest $1,000 each year, for five years, and our portfolio is now worth $6,523.33 (the last $1K investment having been made one year ago). We start our investment portfolio with $P0 and
>Personal? >Okay, I get the idea. This MONTHLY return, R, must satisfy: (1) f(R) = P0(1+R)T + A1(1+R)T1 + A2(1+R)T2 + A3(1+R)T3 + ... + AN(1+R)TN - P = 0 where the original amount, P0, has been invested for T months and the first investment, A1, for T1 months and the next investment, A2, for T2 months and the next investment ... >Okay, I get the idea ... and I assume that an "A" is negative if
that amount is withdrawn. We have a linear approximation for R, namely: (2) RLinear = { P - P0 -Σ An} / {P0 + ΣAnTn } obtained from (1) by replacing things like (1+R)T by the linear approximation 1+TR. To keep things neat, we'll relabel (1+R), calling it x, so we must solve, in place of (1): (3) F(x) = P0xT + A1xT1 + A2xT2 + A3xT3 + ... + ANxTN - P = P0xT + ΣAnxTn - P = 0 where our linear approximation, a la Equation (2), now looks like: (4) x0 = 1 + RLinear = 1 + { P - P0 -Σ An} / {P0 + ΣAnTn } Now we do a single Newton Iteration, starting with x0 and we get - voila!:
>What! I prefer Modified Dietz! >What about pictures? I like ...
>It looks real bad!
This Exact and Newton annualized returns are independent of the values of the portfolio after each transaction. However, if we invent portfolio values - and that's what the spreadsheet does - we can compute a Return rate for each of the ten transactions and get an annualized Modified-Dietz-type Return. Would you like to play with the spreadsheet? The numbers are generated randomly and every F9 re-calculation gives a new set and it's great fun and ...
>zzz ZZZ Of course, you could just use XIRR, right? >zzz ZZZ
|