From: Bob Hanlon on

The form would be

ba = Array[b, 5];

cons = Join[
Thread[Abs[ba] < 1],
Thread[Rest[ba] <= Most[ba]]]

{Abs[b[1]] < 1, Abs[b[2]] < 1, Abs[b[3]] < 1, Abs[b[4]] < 1,
Abs[b[5]] < 1, b[2] <= b[1], b[3] <= b[2], b[4] <= b[3],
b[5] <= b[4]}

params = Join[{a}, ba]

{a, b[1], b[2], b[3], b[4], b[5]}


Bob Hanlon

---- Vindhya Mishra <vindhya.mishra(a)gmail.com> wrote:

=============
Hi
I have a problem with specifying constrants and initial values for
FindFit, when the parameters are defined as a table.

This is my model and the parameters (params)

model = a*(.5 + Sum[b[i]*Cos[2.00000*i*t*Pi/180], {i, 1, 500}]);
params = Join[{a}, Table[b[i], {i, 1, 500}]]
B[i_] = FindFit[data, model, params, t, MaxIterations -> 1000, AccuracyGoal
-> 4]

However I would like to constrain the b[i] parameters to be less than one.
Also if possible , that b[n+1]<=b[n]. What should the syntax be for this?
The number of parameters is too large to specify the constrains
individually.

Thanks,

VM


 | 
Pages: 1
Prev: CellChangeTimes?
Next: Epidemic Model