> FWIW I find the following formulation seems to yield a
> sensible result with Maxima built from CVS (didn't try it
> on any released version).
I have checked the procedure suggested by Robert with the following
setting:
Maxima version: 5.18.1
Maxima build date: 0:30 5/23/2009
host type: i686-pc-linux-gnu
lisp-implementation-type: CLISP
lisp-implementation-version: 2.41 (2006-10-13) (built 3380066971)
(memory 3452045430)
(please, note that my build of CLISP is really old). The error still
persists:
(%i1) M : matrix
([1900, 75.995],
[1910, 91.972],
[1920, 105.711],
[1930, 123.203],
[1940, 131.669],
[1950, 150.697],
[1960, 179.323],
[1970, 203.212],
[1980, 226.505],
[1990, 249.633],
[2000, 281.422]);
(%i2) load (lsquares);
(%i3) mse : lsquares_mse (M, [x, y], y = a * exp(b*(x - 1950)));
(%i4) est : lsquares_estimates_approximate (mse, [a, b], initial=[100,0.01]); *************************************************
N= 2 NUMBER OF CORRECTIONS=25
INITIAL VALUES
F= 2.443742861233416D+42 GNORM= 2.444231559933393D+44
*************************************************
I NFN FUNC GNORM STEPLENGTH
Maxima encountered a Lisp error:
NTH: T is not a list
Automatically continuing.
To reenable the Lisp debugger set *debugger-hook* to nil.
As Leo noted there is a workaround for this, so it is not that bad, but
I'm glad to hear that it can be fixed.
Best regards,
___________________________________
Jos? Antonio Vallejo Rodr?guez
Departamento de Matem?ticas
Facultad de Ciencias UASLP
Lat. Av. Salvador Nava s/n
78290 San Luis Potos? (M?xico)
http://galia.fc.uaslp.mx/~jvallejo/
email: jvallejo at fciencias.uaslp.mx
___________________________________