help

*Subject*: [mg3146] help*From*: OGLAIZOT at ac.dal.ca ("NAME \"Olivier Glaizot\"")*Date*: 7 Feb 1996 05:01:22 -0600*Approved*: usenet@wri.com*Distribution*: local*Newsgroups*: wri.mathgroup*Organization*: Wolfram Research, Inc.*Sender*: daemon at wri.com

I have a problem and maybe some of you can help me with it: I use a recursive function, let say f[x_]:=f[x]=g[f[x-1],y]; with f[0]=0 as a terminal function. Now within the function g, there is a maximum to be found, something like Max[f[x-1],{y,0,ysup}] y being a discrete parameter (y=0,1,...ysup) It is very time consuming, but I know that the discrete function to be maximized has always a maximum, usually for small values of y. Then, my idea was instead of building the whole Table {g[f[x-1],[y1],... g[f[x-1],ysup], to stop as soon as g decrease again, with a function of the type While[g[yi]>g[yi-1],i++,temp=g[yi]], so the computation stops as soon as the function g decreases. Now the problem is that as it is a recursive function, the While[] is called in a nested way, and the initial values of i reset each time, at least I think it is what happens, and the results are of course wrong. Does anyone has a good idea? Or is it not clear (I'm afraid so...) thanks, Olivier ____________________________________________________________________________ | OLIVIER GLAIZOT ****NEW ADDRESS FROM THE FIRST MAY 1996**** | | Biology Department Institute of Zoology and Animal Ecology | | Dalhousie University Biology Building, University of Lausanne | | Halifax, N.S. CH-1015 DORIGNY | | B3H 4J1 CANADA SWITZERLAND | | e-mail: oglaizot at ac.dal.ca olivier.glaizot at izea.unil.ch | |_________________________________________________________________________ | | Home address: | | 1523 Chesnut Street | | Halifax B3H 3S9 | | CANADA | | Tel/Fax: (1-902) 423-9397 | |___________________________________________________________________________|