http://www.mathworks.com/matlabcentral/newsreader/view_thread/317945
MATLAB Central Newsreader  lsqcurvefit and integral
Feed for thread: lsqcurvefit and integral
enus
©19942015 by MathWorks, Inc.
webmaster@mathworks.com
MATLAB Central Newsreader
http://blogs.law.harvard.edu/tech/rss
60
MathWorks
http://www.mathworks.com/images/membrane_icon.gif

Tue, 13 Mar 2012 20:07:34 +0000
lsqcurvefit and integral
http://www.mathworks.com/matlabcentral/newsreader/view_thread/317945#869860
pmassicotte
Hi everyone.<br>
<br>
I have a set of differential equations of which one is in the<br>
following form:<br>
<br>
dx/dt = F1 + F2  int(F3,a,b)<br>
<br>
I would like to use lsqcurvefit to find optimal parameters (between<br>
specified boundaries). However, I never used lsqcurvefit with an<br>
integral. I was wondering if it was doable or if I should be aware of<br>
anything before going forward.<br>
<br>
I hope I have been clear enough in my question.<br>
<br>
Regards,<br>
Phil

Wed, 14 Mar 2012 02:41:19 +0000
Re: lsqcurvefit and integral
http://www.mathworks.com/matlabcentral/newsreader/view_thread/317945#869889
Roger Stafford
pmassicotte <pmassicotte@hotmail.com> wrote in message <96580354d8a74309a2cafaf6924b23f8@i2g2000vbv.googlegroups.com>...<br>
> ......<br>
> I have a set of differential equations of which one is in the<br>
> following form:<br>
> <br>
> dx/dt = F1 + F2  int(F3,a,b)<br>
> .......<br>
      <br>
Do F3, a, or b vary with t? I assume F3 also has its own variable of integration independent of t that ranges from a to b.<br>
<br>
Roger Stafford

Wed, 14 Mar 2012 12:00:51 +0000
Re: lsqcurvefit and integral
http://www.mathworks.com/matlabcentral/newsreader/view_thread/317945#869938
pmassicotte
On Mar 13, 10:41 pm, "Roger Stafford"<br>
<ellieandrogerxy...@mindspring.com.invalid> wrote:<br>
> pmassicotte <pmassico...@hotmail.com> wrote in message <96580354d8a74309a2cafaf6924b2...@i2g2000vbv.googlegroups.com>...<br>
> > ......<br>
> > I have a set of differential equations of which one is in the<br>
> > following form:<br>
><br>
> > dx/dt = F1 + F2  int(F3,a,b)<br>
> > .......<br>
><br>
>       <br>
> Do F3, a, or b vary with t? I assume F3 also has its own variable of integration independent of t that ranges from a to b.<br>
><br>
> Roger Stafford<br>
<br>
Hi there.<br>
<br>
a and b and independent of t. Your assumption is also correct.<br>
<br>
Regards,<br>
Phil

Wed, 14 Mar 2012 12:11:56 +0000
Re: lsqcurvefit and integral
http://www.mathworks.com/matlabcentral/newsreader/view_thread/317945#869941
Torsten
Am Mittwoch, 14. März 2012 13:00:51 UTC+1 schrieb pmassicotte:<br>
> On Mar 13, 10:41 pm, "Roger Stafford"<br>
> <ellieandrogerxy...@mindspring.com.invalid> wrote:<br>
> > pmassicotte <br>
> wrote in message <96580354d8a74309a2cafaf6924b2...@i2g2000vbv.googlegroups.com>...<br>
> > > ......<br>
> > > I have a set of differential equations of which one is in the<br>
> > > following form:<br>
> ><br>
> > > dx/dt = F1 + F2  int(F3,a,b)<br>
> > > .......<br>
> ><br>
> >       <br>
> > Do F3, a, or b vary with t? I assume F3 also has its own variable of integration independent of t that ranges from a to b.<br>
> ><br>
> > Roger Stafford<br>
> <br>
> Hi there.<br>
> <br>
> a and b and independent of t. Your assumption is also correct.<br>
> <br>
> Regards,<br>
> Phil<br>
<br>
Is your equation already a part of a discretized partial differential equation ?<br>
Does F3 depend on x ? <br>
If yes: a and b are (fixed) time instances ?<br>
If no: how do you get x for values in between a and b ?<br>
<br>
Please explain your problem in more detail.<br>
<br>
Best wishes<br>
Torsten.

Wed, 14 Mar 2012 13:06:59 +0000
Re: lsqcurvefit and integral
http://www.mathworks.com/matlabcentral/newsreader/view_thread/317945#869945
pmassicotte
On Tuesday, March 13, 2012 4:07:34 PM UTC4, pmassicotte wrote:<br>
> Hi everyone.<br>
> <br>
> I have a set of differential equations of which one is in the<br>
> following form:<br>
> <br>
> dx/dt = F1 + F2  int(F3,a,b)<br>
> <br>
> I would like to use lsqcurvefit to find optimal parameters (between<br>
> specified boundaries). However, I never used lsqcurvefit with an<br>
> integral. I was wondering if it was doable or if I should be aware of<br>
> anything before going forward.<br>
> <br>
> I hope I have been clear enough in my question.<br>
> <br>
> Regards,<br>
> Phil<br>
<br>
fdg<br>
<br>
On Tuesday, March 13, 2012 4:07:34 PM UTC4, pmassicotte wrote:<br>
> Hi everyone.<br>
> <br>
> I have a set of differential equations of which one is in the<br>
> following form:<br>
> <br>
> dx/dt = F1 + F2  int(F3,a,b)<br>
> <br>
> I would like to use lsqcurvefit to find optimal parameters (between<br>
> specified boundaries). However, I never used lsqcurvefit with an<br>
> integral. I was wondering if it was doable or if I should be aware of<br>
> anything before going forward.<br>
> <br>
> I hope I have been clear enough in my question.<br>
> <br>
> Regards,<br>
> Phil<br>
<br>
<br>
<br>
On Tuesday, March 13, 2012 4:07:34 PM UTC4, pmassicotte wrote:<br>
> Hi everyone.<br>
> <br>
> I have a set of differential equations of which one is in the<br>
> following form:<br>
> <br>
> dx/dt = F1 + F2  int(F3,a,b)<br>
> <br>
> I would like to use lsqcurvefit to find optimal parameters (between<br>
> specified boundaries). However, I never used lsqcurvefit with an<br>
> integral. I was wondering if it was doable or if I should be aware of<br>
> anything before going forward.<br>
> <br>
> I hope I have been clear enough in my question.<br>
> <br>
> Regards,<br>
> Phil<br>
<br>
<br>
<br>
On Tuesday, March 13, 2012 4:07:34 PM UTC4, pmassicotte wrote:<br>
> Hi everyone.<br>
> <br>
> I have a set of differential equations of which one is in the<br>
> following form:<br>
> <br>
> dx/dt = F1 + F2  int(F3,a,b)<br>
> <br>
> I would like to use lsqcurvefit to find optimal parameters (between<br>
> specified boundaries). However, I never used lsqcurvefit with an<br>
> integral. I was wondering if it was doable or if I should be aware of<br>
> anything before going forward.<br>
> <br>
> I hope I have been clear enough in my question.<br>
> <br>
> Regards,<br>
> Phil<br>
<br>
<br>
<br>
On Tuesday, March 13, 2012 4:07:34 PM UTC4, pmassicotte wrote:<br>
> Hi everyone.<br>
> <br>
> I have a set of differential equations of which one is in the<br>
> following form:<br>
> <br>
> dx/dt = F1 + F2  int(F3,a,b)<br>
> <br>
> I would like to use lsqcurvefit to find optimal parameters (between<br>
> specified boundaries). However, I never used lsqcurvefit with an<br>
> integral. I was wondering if it was doable or if I should be aware of<br>
> anything before going forward.<br>
> <br>
> I hope I have been clear enough in my question.<br>
> <br>
> Regards,<br>
> Phil

Wed, 14 Mar 2012 13:06:39 +0000
Re: lsqcurvefit and integral
http://www.mathworks.com/matlabcentral/newsreader/view_thread/317945#869946
pmassicotte
On Wednesday, March 14, 2012 8:11:56 AM UTC4, Torsten wrote:<br>
> Am Mittwoch, 14. März 2012 13:00:51 UTC+1 schrieb pmassicotte:<br>
> > On Mar 13, 10:41 pm, "Roger Stafford"<br>
> > <ellieandrogerxy...@mindspring.com.invalid> wrote:<br>
> > > pmassicotte <br>
> > wrote in message <96580354d8a74309a2cafaf6924b2...@i2g2000vbv.googlegroups.com>...<br>
> > > > ......<br>
> > > > I have a set of differential equations of which one is in the<br>
> > > > following form:<br>
> > ><br>
> > > > dx/dt = F1 + F2  int(F3,a,b)<br>
> > > > .......<br>
> > ><br>
> > >       <br>
> > > Do F3, a, or b vary with t? I assume F3 also has its own variable of integration independent of t that ranges from a to b.<br>
> > ><br>
> > > Roger Stafford<br>
> > <br>
> > Hi there.<br>
> > <br>
> > a and b and independent of t. Your assumption is also correct.<br>
> > <br>
> > Regards,<br>
> > Phil<br>
> <br>
> Is your equation already a part of a discretized partial differential equation ?<br>
> Does F3 depend on x ? <br>
> If yes: a and b are (fixed) time instances ?<br>
> If no: how do you get x for values in between a and b ?<br>
> <br>
> Please explain your problem in more detail.<br>
> <br>
> Best wishes<br>
> Torsten.<br>
<br>
Hi there. Thank you for your answer.<br>
<br>
F3 does not depend on x. For instance, F3 is a function that takes values in a table. In fact it was my next question. How to calculate integral that takes its values in a table (measured values). In my case, F3 is:<br>
<br>
<a href="http://i44.tinypic.com/300cheo.png">http://i44.tinypic.com/300cheo.png</a><br>
<br>
where for example Ed(0,lambda) is the light intensity measured at lambda wavelength. <br>
<br>
Once again, I hope I'm clear enough. Don't hesitate to ask me more question. <br>
<br>
Your help is very valuable.<br>
<br>
Regards,<br>
Phil