On Fri, 24 Nov 2006, Miquel Cabanas wrote:
> after checking some previously posted messages, it seems that a better
> way to achieve your goal is to use a function definition,
>
> (%i1) display2d : false;
>
> (%o1) FALSE
> (%i2) f(x) :=
> if x=1 then 1/(1+s)
> else (1+f(x-1)*s)/(1+s*(1+f(x-1)));
>
> (%o2) f(x):=IF x = 1 THEN 1/(1+s) ELSE (1+f(x-1)*s)/(1+s*(1+f(x-1)))
> (%i3) f(1);
>
> (%o3) 1/(s+1)
> (%i4) f(2);
>
> (%o4) (s/(s+1)+1)/(s*(1/(s+1)+1)+1)
> (%i5) radcan(f(2));
>
> (%o5) (2*s+1)/(s^2+3*s+1)
> (%i6) radcan(f(10));
>
> (%o6) (10*s^9+165*s^8+792*s^7+1716*s^6+2002*s^5+1365*s^4+560*s^3+136*s^2
> +18*s
> +1)
> /(s^10+55*s^9+495*s^8+1716*s^7+3003*s^6+3003*s^5+1820*s^4+680*s^3
> +153*s^2+19*s+1)
>
>
> Miquel
>
Thank you very much, this works quite nicely indeed.
I'm more and more impressed by the power of Maxima.
regards,
hugo