A
Augabog
I'm using SOLVER to change three variables in order to get a minimum
sum squared error. I've noticed that if I change the initial values
for the 3 variables before running solver, that the solver will return
different values! There should only be one value for each associated
with a minimum sum of squared error, and I have placed non constraints
on what Excel can change the three variables in order to reach that
minimum value for the target cell (e.g. the sum squared error value).
Therefore, what the values for those three cells are prior to running
the solver shouldn't matter, even if they say nothing at all. This
concerns me about the accuracy of SOLVER. Has anyone else had this
issue? If so, is there an agreed upon way to set up the initial
variables prior to running SOLVER?
Please let me know if I need to elaborate.
Thanks,
Hosley
sum squared error. I've noticed that if I change the initial values
for the 3 variables before running solver, that the solver will return
different values! There should only be one value for each associated
with a minimum sum of squared error, and I have placed non constraints
on what Excel can change the three variables in order to reach that
minimum value for the target cell (e.g. the sum squared error value).
Therefore, what the values for those three cells are prior to running
the solver shouldn't matter, even if they say nothing at all. This
concerns me about the accuracy of SOLVER. Has anyone else had this
issue? If so, is there an agreed upon way to set up the initial
variables prior to running SOLVER?
Please let me know if I need to elaborate.
Thanks,
Hosley