MATLAB linear fit problem

2) Modify your linear least squared code so that it can fit data to a power function and plot both the data as scatter and the best fit curve as a smooth function on the same set of axes. Submit the modified code and, once it is working, a graph of period versus orbital radius for the Jovian moons. Write out/print the best fit power function equation that your code generated.

This is the question and I have my data in a file called ' moon_data', I try and run the data through a script file I have for finding the linear fit called 'get_linear_fit_2".

I keep getting a "too many output arguments" error when it runs the data thru my cramers rule script that I have already used many times in the past. I can't seem to figure out how this is, can somebody help me?

function fit= get_linear_fit_2(data)
s = size(data)
Syx=0
Sxx=0
Sy=0
Sx=0
for n=1:s(1)
    Syx=data(n,1)*data(n,2)+Syx
    Sxx=(data(n,1))^2+sxx
    Sx=data(n,1)+Sx
    Sy=data(n,2)+Sy
end
m=[Sxx,Sx;Sx,s(1)]
c=[Syx;Sy]
fit=Cramers_rule(m,c)
end
%Linear fit data above
function Cramers_rule(m,c)
s=size(m)
n=1
dm=det(m)
while s(2)>=n
    a=m
    a(:,n)=c(:)
    y(n)=det(a)/dm
    disp(y(n))
    n=n+1
end
end
%Cramers Rule function above

My data code is just the two data sets in a script file alone, thats it.

Please help!!

Answer 1

[Image Analyst]

You say

fit=Cramers_rule(m,c)

as if Cramers_rule() returns something. You can see from your function declaration:

function Cramers_rule(m,c)

that is does NOT return anything, so there is nothing that can be assigned to your "fit" variable. Hence the error message you got.

[Erin Szaruga]

Okay that makes sense to me but what do I change about my code that would make it so MATLAB runs this thru and returns a fit for a power function?