DDAYS in the solver

[Bvr]

Hi,
I try to use the DDAYS command in the solver but get always "Bad Guess(es):
E1: A=DDAYS(D1;D2)

D1:2006,1010
D2:2006,1111

A: 1,0000 as seed

Any clue?
Thx.
Kr.,
Bvr

[Tim Wessman]

(10-21-2016 07:46 AM)Bvr Wrote:  Hi,
I try to use the DDAYS command in the solver but get always "Bad Guess(es):
E1: A=DDAYS(D1;D2)

D1:2006,1010
D2:2006,1111

A: 1,0000 as seed

Any clue?
Thx.
Kr.,
Bvr

I'm guessing this would not work in the solver due to it not being a continuous function with no derivative. I'm pretty sure the same limitation would have been in the 48 series too.

[Tim Wessman]

(10-21-2016 04:32 PM)moonbeam Wrote:  If the arguments of DDAYS are given, the function value is a constant. Solver can handle this just fine :-) Looking for the start or end date is another matter, best left to DateAdd(date,ndays).

Yes, you are right here of course. I was specifically thinking the "solve for any of the variables" aspect part of the solver like you point out. Thanks for the clarification!

[Bvr]

Thanks guys,
I used Dstart and Dend and it works perfect now.

Kr.,
Bvr