Re: Curve fit

[Moo]

On Sat, Nov 20, 2010 at 7:50 PM, cunninghands <address@hidden> wrote:

I'm starting to understand it at last!

We just need to find the solution to the equation that would best fit the
data points.

Liam, thanks a lot.

 

Well this is more of a question of what exactly you're trying to do.  Theoretically you could make an equation that would go through all the points exactly, but the equation will be useless to analyze.  You probably don't want this.  

Since the original equation were data from a physical process (binding energy per nucleon), there is probably a theoretical curve that you (or your teacher) are expecting the points to follow.  This, while it may not be the "best" fit to the data points, is the one that you should be using.  The goal in experiment is usually to explain possible causes for why you data is off from the expected curve, and maybe to explain the "best" coefficients that fit the points.

More than likely at some point the teacher gave you (or derived for you) the theoretical equation with coefficients.  This is what you want to fit your data against.