41C/CV root finders

[Ángel Martin]

(05-29-2015 06:15 PM)Dieter Wrote:  

(05-29-2015 08:04 AM)Ángel Martin Wrote:  f(i) = PV + (1+ip) PMT [ (1- (1+i)^-n) / i] + FV (1+i)^-n
f ’ (i) = (PMT / i^2 ) * [ (1+i)^(-n) - 1 ] + n * [PMT (1 + ip)/i – FV ] * (1+i)^-(n+1)

where p=1 in Begin mode or 0 in End mode.

The formulas look a bit complicated, and I hope they are implemented efficiently, but anyway: they seem to be correct.

I think they are correct and correctly implemented now after I fixed the bug - see previous post. - but you're welcome to improve the method any time ;-)

(05-29-2015 06:15 PM)Dieter Wrote:  So in all cases 4 – 6 iterations will do, depending on the initial estimate.

That's why I wonder why 7 to 13 iterations are required before the solver in the TVM ROM converges. This sounds like an error in the implementation. Maybe you can compare the results of every iteration step with the ones in the tables above.

You may have missed my previous post, or perhaps our posts crossed paths...
it is down to 4 or 5 iterations now after I fixed the bug I found in the algorithm.

Cheers,
'AM