Devon McCormick wrote: > I am not making this assumption at all. > > On 12/9/06, Chris Burke <[EMAIL PROTECTED]> wrote: >> >> The mistake is the assumption that: >> >> (x^r)+(y^r) is the same as (x+y)^r >> >> i.e. >> >> */365 $ +/0.6 0.4 * 3 2 ^ %365 >> 2.55099 >> >> */365 $ (+/0.6 0.4 * 3 2) ^ %365 >> 2.6
Nevertheless, there is a mistake in the arithmetic. You need to average (i.e. weighted average) the annual rates before calculating the daily rate, and not average the daily rates. This explains the observed discrepancy when calculating the annual rate from the daily rate. Another way to see this is, is to calculate the annual rate by using power - this at least should be accurate, even if the times reduce on 365 values is inaccurate. Yet this returns exactly the values shown earlier. (+/0.6 0.4 * 3 2 ^ %365) ^ 365 2.55099 ((+/0.6 0.4 * 3 2) ^ %365) ^ 365 2.6 Perhaps the closeness of the two results suggests rounding errors - but try larger numbers to see that the difference has to be something other than rounding errors. ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
