Solver in Excel97

[Bill Martin]

I've used Solver in Excel97 with no problems in the past. Today it's not
working properly. I'm trying to maximize one cell based on the contents of
one other cell. Essentially Solver always just comes back without doing any
iterations and says it's found the solution -- and just returns the initial
starting value. Hmmm....

To test I created a little function within the same work sheet that is the
sum of two offset Normal curves and ask solver to find the peak. It always
works fine -- though it will of course also find the wrong peak depending on
the starting point.

Ok, maybe the problem is that my real data is discontinuous. So I took my
Normal curve data and copied the values only over to a table and created a
VLOOKUP function to search for the optimum X which returns the maximum Y from
the table. Solver still works fine.

Yet with my real data it says it worked but doesn't actually even try. I
tell it to pause at each iteration so I can follow what's going on, and the
Solver doesn't iterate at all with the real data. Yet with the Normal data
it does.

Does any of this sound familiar? I wish the Solver also had a Monte Carlo
solution method built into it.

Thanks....

Bill -- (Remove KILLSPAM from my address to use it)

 

[Ron Rosenfeld]

I've used Solver in Excel97 with no problems in the past. Today it's not
working properly. I'm trying to maximize one cell based on the contents of
one other cell. Essentially Solver always just comes back without doing any
iterations and says it's found the solution -- and just returns the initial
starting value. Hmmm....

To test I created a little function within the same work sheet that is the
sum of two offset Normal curves and ask solver to find the peak. It always
works fine -- though it will of course also find the wrong peak depending on
the starting point.

Ok, maybe the problem is that my real data is discontinuous. So I took my
Normal curve data and copied the values only over to a table and created a
VLOOKUP function to search for the optimum X which returns the maximum Y from
the table. Solver still works fine.

Yet with my real data it says it worked but doesn't actually even try. I
tell it to pause at each iteration so I can follow what's going on, and the
Solver doesn't iterate at all with the real data. Yet with the Normal data
it does.

Does any of this sound familiar? I wish the Solver also had a Monte Carlo
solution method built into it.

Thanks....

Bill -- (Remove KILLSPAM from my address to use it)

There are some bugs in Solver for XL 97 and I had some worksheets on which
Solver would not work. I don't recall all of the details.

I have a hazy recollection that merged cells could cause a problem; as could
deleted NAMES and there may even be a problem with NAMES on multiple
worksheets.

Later versions of Excel have improved the Solver. There may have been an
updated Solver released for XL97, but you'd have to check the MSKB for that.


--ron

 

[Bill Martin]

There are some bugs in Solver for XL 97 and I had some worksheets on which

Solver would not work. I don't recall all of the details.

I have a hazy recollection that merged cells could cause a problem; as could
deleted NAMES and there may even be a problem with NAMES on multiple
worksheets.

Later versions of Excel have improved the Solver. There may have been an
updated Solver released for XL97, but you'd have to check the MSKB for that.

Thanks Ron....

Bill -- (Remove KILLSPAM from my address to use it)