Hello,
Can you please help me with finding least squares estimates for the
following data and function:
(%i18) grind(dm);
matrix([4,157862744],[6,81514144],[8,42899496],[10,20773720],[12,2699392],
[14,0],[3,244256672],[4,153504224],[5,150344088],[6,83708024],
[7,138026400],[8,56704496],[9,33277704],[10,36540352],[11,65270544],
[12,18357256],[13,63192616],[14,29463776],[15,6964280],[16,673976])$
(%o18) done
(%i19) mse : lsquares_mse(dm, [x,y], a*exp(-b*x) = y);
20
==== - b dm
\ i, 1 2
> (a %e - dm )
/ i, 2
====
i = 1
(%o19) --------------------------------
20
(%i20) lsquares_estimates_approximate(mse, [a,b], initial = [ 10^9, 0.5 ],
iprint = [ -1, 0 ]);
IFLAG= -1
LINE SEARCH FAILED. SEE DOCUMENTATION OF ROUTINE MCSRCH
ERROR RETURN OF LINE SEARCH: INFO= 3
POSSIBLE CAUSES: FUNCTION OR GRADIENT ARE INCORRECT
OR INCORRECT TOLERANCES
(%o20) [[]]
I can see by plotting that the initial parameters hit the data pretty
well. Can you give me a clue why it fails and what I could do to correct
it? Also, what are the meanings of the undocumented "iprint" parameter and
of the "tol" parameter, which is mentioned, but not explained in the
documentation?
Regards,
Jan Ploski