How to set a parameter of poly2 fit to a specific value during fit process?

I'm trying to take data and plot it then fit a curve that is of the form f(x) = p1*x^2 + p2*x + p3, but with p2 equal to zero. Right now my code fits for p1, p2, and p3. How can I set p2 to zero and allow it only to fit p1 and p3? Please see the code and results below. Thank you!

Code:

rho=Resistivityohmcm;
T=TemperatureK;
plot(T,rho,'-o')
f=fit(T,rho,'poly2');
plot(f,T,rho)
disp (f)

Results: 
Gives the correct plot and the values for f:
Linear model Poly2:

       f(x) = p1*x^2 + p2*x + p3
       Coefficients (with 95% confidence bounds):
         p1 =   1.511e-07  (1.422e-07, 1.599e-07)
         p2 =  -2.127e-05  (-2.512e-05, -1.743e-05)
         p3 =    0.001779  (0.001383, 0.002175)

Answer 1

[dpb]

You have to define the model...for something this simple the anonymous function route is the simplest...

fnPolySq = @(p1,p2,x) p1*x.^2 + p2;
f=fit(T,rho,'fnPolySq');

ADDENDUM

It's the pits getting old...forgot where I was going in mid-stream... :(

The above estimates the function without the linear term; the Q? was for a constant value...same idea, build it into the model

C=YourValue;
fnPolySq = @(p1,p3,x) p1*x.^2 + C*x + p3;

[Alexandria Will-Cole]

I tried to implement this and the following code is below... It gives me an error for using this fit function. Any insight? Thanks.

Code:

rho=Resistivityohmcm; T=TemperatureK;

plot(T,rho,'-o')

fnPolySq = @(p1,p2,x,y) p1*x.^2 + p2; f=fit(T,rho,'fnPpolySq');

plot(f,T,rho) disp (f)

Error:

Error using fittype>iCreateFromLibrary (line 412) Library function fnPpolySq not found.

Error in fittype>iCreateFittype (line 343) obj = iCreateFromLibrary( obj, varargin{:} );

Error in fittype (line 328) obj = iCreateFittype( obj, varargin{:} );

Error in fit>iFit (line 157) model = fittype( fittypeobj, 'numindep', size( xdatain, 2 ) );

Error in fit (line 108) [fitobj, goodness, output, convmsg] = iFit( xdatain, ydatain, fittypeobj, ...

[Alexandria Will-Cole]

Also, I corrected the typo for the "Ppoly" and still get an error.

[dpb]

Oops...sorry about the typo. I see I also inadvertently threw in the y variable in the argument list for the anonymous function...that doesn't belong there.

Updated the Answer to correct...I did a sample with some data just happened to have loaded here...wasn't a great data set, but works to show principle--

>> fnPolySq=@(p1,p2,x) p1*x.^2 + p2;
>> fit(x,y,fnPolySq)
Warning: Start point not provided, choosing random start point. 
> In curvefit.attention.Warning/throw (line 30)
  In fit>iFit (line 299)
  In fit (line 108) 
ans = 
     General model:
     ans(x) = p1*x.^2+p2
     Coefficients (with 95% confidence bounds):
       p1 =    0.002188  (0.001458, 0.002918)
       p2 =       2.988  (2.862, 3.114)
>> f=ans;
>> plot(f,x,y)
>>

[Alexandria Will-Cole]

Thank you! This works very well! I appreciate it!

[dpb]

You could then Accept answer to indicate topic closed if nothing else...

Glad to help...