Predictor and Response Variables must have same length error

Hello,

I am attempting to use fitnlm, and come across an error I don't understand.

x =[p;l];
model=@(b,x)((x(1,:)+x(2,:)+b(2))-sqrt((x(1,:)+x(2,:)+b(2)).^2)-(4*x(1,:)*x(2,:)))*(b(1)/(2*x(1,:)));
beta0=[0.1,1];
for i=1:rows
    y=csp(i,:)
    x
    size(x(1,:))
    size(x(2,:))
    size(y)
  fitnlm(x,y,model,beta0)
end

Issue is, the x and y are the same length so I don't quite know why I'm getting this error

y =
     0     0     0     0     0     0     0     0     0
x =

    0.1500    0.1800    0.1900    0.2100    0.2300    0.2500    0.2600    0.2800    0.3000
    1.1000    0.8800    0.7700    0.6300    0.4700    0.3700    0.2600    0.1400         0
ans =

     1     9
ans =

     1     9
ans =

     1     9

Answer 1

[the cyclist]

fitnlm expects column vectors as input, not row vectors. Rows are the observations, and columns are variables.

y = [0     0     0     0     0     0     0     0     0]';

x = [0.1500    0.1800    0.1900    0.2100    0.2300    0.2500    0.2600    0.2800    0.3000;
     1.1000    0.8800    0.7700    0.6300    0.4700    0.3700    0.2600    0.1400         0]';

model=@(b,x)((x(:,1)+x(:,2)+b(2))-sqrt((x(:,1)+x(:,2)+b(2)).^2)-(4*x(:,1).*x(:,2))).*(b(1)./(2.*x(:,1)));
beta0=[0.1,1];

fitnlm(x,y,model,beta0)

Warning: Rank deficient, rank = 1, tol = 9.077909e-15.

Warning: Rank deficient, rank = 1, tol = 9.037373e-15.

Warning: Rank deficient, rank = 1, tol = 9.033309e-15.

Warning: Rank deficient, rank = 1, tol = 9.032903e-15.

Warning: Some columns of the Jacobian are effectively zero at the solution, indicating that the model is insensitive to some of its parameters. That may be because those parameters are not present in the model, or otherwise do not affect the predicted values. It may also be due to numerical underflow in the model function, which can sometimes be avoided by choosing better initial parameter values, or by rescaling or recentering. Parameter estimates may be unreliable.

ans = 
Nonlinear regression model:
    y ~ ((x1 + x2 + b2) - sqrt((x1 + x2 + b2)^2) - (4*x1*x2))*(b1/(2*x1))

Estimated Coefficients:
          Estimate        SE        tStat      pValue 
          ________    __________    ______    ________

    b1    9.89e-16    3.4966e-16    2.8284    0.022204
    b2           1             0       Inf           0


Number of observations: 9, Error degrees of freedom: 8
Root Mean Squared Error: 1.29e-15
R-Squared: -Inf,  Adjusted R-Squared -Inf
F-statistic vs. zero model: 0, p-value = 1

Note that I also did two fixes to your model definition:

• converted it to accept the columns instead of the rows
• used element-wise multiplication and division operators, instead of matrix operations